The ultimate goal of radiation therapy (RT) is to deliver enough radiation to eradicate all tumor clonogens within the irradiated field while also minimizing dose to adjacent normal structures so as to cause no (or minimal) normal tissue injury. With few exceptions, the dose of radiation currently delivered to most anatomic sites is limited by the tolerance of these adjacent normal organs, thereby restricting the dose of radiation deposited in tumors to an amount that may or may not be ideal to realize an optimal cure. The goal of many advanced technologies introduced into RT are aimed at addressing this problem.
The ultimate goal of radiation therapy (RT) is to deliver enough radiation to eradicate all tumor clonogens within the irradiated field while also minimizing dose to adjacent normal structures so as to cause no (or minimal) normal tissue injury. With few exceptions, the dose of radiation currently delivered to most anatomic sites is limited by the tolerance of these adjacent normal organs, thereby restricting the dose of radiation deposited in tumors to an amount that may or may not be ideal to realize an optimal cure. The goal of many advanced technologies introduced into RT are aimed at addressing this problem. While technologic improvements have occurred in all aspects of RT (radiobiology, drug delivery, physics, etc) in an attempt to maximize this therapeutic ratio, those related to the imaging, planning, and delivery of externally applied radiation have a direct impact on the feasibility and success of RT and are considered by Vikram et al in the accompanying article.[1]
When and how to appropriately introduce new advanced technologies into medical practice is highly controversial and a topic of much debate. We laud Dr. Vikram and his colleagues for starting an organized discussion of this topic in RT. The path followed by Vikram et al is relatively straightforward: (1) describe the current results of conventional RT with particular reference to dose-limiting toxicity, (2) summarize the benefit observed from advanced technologies over standard techniques from randomized clinical trials (RCTs), and (3) determine the opportunities and priorities for RCTs and areas requiring further technologic development.
Clinical Equipoise
The paucity of RCTs in cancer research in general has been clearly documented and is certainly lacking in the evaluation of advanced technologies applied in RT.[2] In practice, modern RT involves several sequential steps-beginning with imaging, followed by planning, and culminating in treatment delivery-and most would agree that an RCT is not necessary for each small advanced technology iteration or modification introduced. Perhaps a more important issue regarding the systematic use of RCTs for testing advanced technologies is that clinical equipoise might not always be present. In this context, we suggest that clinical equipoise means there is genuine uncertainty about whether the experimental treatment or advanced technologies would be beneficial.
Some authors argue that because modern technology has been associated with demonstrable theoretical improvements (such as improvements in calculated radiation dose distribution), the principle of equipoise is violated and thus it would be unethical to perform such studies. Others would suggest that individual equipoise is different from collective equipoise, the latter referring to the profession as a whole or “experts” in the field as opposed to the individual clinician. When there is a lack of general consensus on a specific topic (such as the clinical significance of an alteration in radiation dose distribution), an RCT is then ethical and reasonable. Thus, the challenge becomes in defining whether (and to what extent) there is uncertainty regarding the benefits of advanced technologies over standard approaches, and if such benefits exist, whether clinically significant differences can be determined and measured.
Quality Endpoints
One of the most commonly studied aspects in advanced technologies research is the ability of the advanced technology to modify the quality of a given treatment. However, if one cannot clearly define quality and has trouble in reproducibly measuring quality, it is quite unlikely that an RTC of an advanced technology will provide meaningful results. As noted by Vikram et al, a specific focus on defining exact quality endpoint(s) that a given advanced technology should impact and how an alteration in that quality endpoint would result in improved treatment are critical, first steps toward logical RTC design of advanced technologies. In addition, such focus would provide the impetus for the design of tools to measure quality endpoints, allowing uniform assessment across institutions. It would be foolish to assume that every advanced technology–inspired technologic improvement or resultant dosimetric alteration will necessarily result in clinically meaningful superior treatment outcomes. Thus, in order for rigorous clinical trials to test these advances for clinical significance, tool sets that include uniform, accepted measures and metrics of quality must be developed. It would seem reasonable that these quality measures, metrics, and tools could be developed by national radiation oncology and radiation physics consortia in concert with centers of excellence. A distribution of such tool sets to centers that participate in cooperative group studies would then significantly and quickly enhance the ability to study the capacity of the broad radiation oncology community to perform advanced technologies research and to determine with what quality it can be performed. It is likely that such work would help to improve the low sensitivity and specificity of clinical outcomes as an indicator of treatment quality.[3]
It should be stated that there are aspects of advanced technologies that are likely not suitable for strict quality assurance evaluation using standard, predefined endpoints or external review. For example, the definition/delineation of target volumes including the gross tumor volume (GTV) is rarely done in the absence of relevant clinical information, all of which goes into the development of the final GTV. Without intimate knowledge of these important clinical factors, it is very difficult to determine the “correct” GTV. Results of clinical trials testing advanced technologies that are designed to conform radiation dose to the GTV would, therefore, be difficult to interpret. The quality assessment tool sets discussed earlier will have to become quite sophisticated to capture the sometimes subtle differences in tumor volume definitions that may be “correct” but not necessarily “standard.”
Conclusions
The need for improvement in RT is clearly documented by Vikram et al, and the introduction of new advanced technologies may be one effective way to achieve important clinical gains. We believe that the introduction of many of these advanced technologies have outpaced our present clinical expertise to define accurate and agreed-upon treatment volumes and (with few exceptions) have highlighted our inability to precisely define the tolerance of normal structures.
In our view, the first evaluation of advanced technologies should be done within “controlled” studies designed to test agreed-upon anatomic definitions, atlases, and quality measurement instruments, so that the validation of these common tools is completed before embarking on RTCs of advanced technologies. Given the enormity of the cost associated with many advanced technologies, rigorous testing must be done to determine whether improvements in the quality endpoints these technologies are purported to provide actually have a meaningful clinical impact. Validated measurement tools are key in this process.
1. Vikram B, Coleman CN, Deye JA: Current status and future potential of advanced technologies in radiation oncology: Part 2. State of the science by anatomic site. Oncology (Williston Park) 23:380-385, 2009.
2. Hillner BE: Trends in clinical trials reports in common cancers between 1989 and 2000. J Clin Oncol 21:1850-1858, 2003.
3. Bentzen SM: Randomized controlled trials in health technology assessment: Overkill or overdue? Radiother Oncol 86:142-147, 2008.