The goal of the NIH-supported Patient-Reported Outcomes Measurement Information System (PROMIS) is to enhance and standardize the assessment of patient-reported healthrelated quality of life (HRQOL). Started in 2004, the PROMIS Network has utilized a sophisticated and rigorous multimethod approach to develop item banks measuring physical function, pain, fatigue, depression, anxiety, anger, and social well-being. Researchers can access PROMIS item banks through the PROMIS website (www.NIHpromis.org) to select either static or individually tailored adaptive measures of HRQOL. Either approach yields brief, precise, and valid measures of a patient’s health status.
In a special supplement funded by the NCI, the PROMIS network has extended the item banks to capture data on sexual functioning, cognitive function, sleep/wake function, and illness impact in cancer populations. This supplemental project will evaluate how well the PROMIS item banks perform, both in the treatment and the survivorship phases of care.
Why Clinical Trials Are Needed
Hypotheses regarding improved patient outcomes that are based on physical dose distributions and computer-generated treatment plans require appropriate clinical studies to validate those hypotheses. Admittedly very limited proof of principle currently exists, but the absence of proof is not proof of absence of benefit from the advanced technologies, because time is required for generating credible clinical trial data—both short- and long-term—on survival and patient-reported outcomes. Nonetheless, important and informative data on acute toxicities can be gathered fairly quickly.
Quality assurance procedures that can meet the demands of the advanced technologies must be established and implemented in conjunction with undertaking formal comparative studies. For instance, atlases for delineating target volumes and organs at risk, as well as common tools for standardization of image registration (fusion) software, have been developed as already described. Methods for dealing with the deformation of the target volumes and organs at risk, procedures for preventing errors in treatment delivery due to movement during and between fractions of radiation, and procedures for ensuring calculation of the correct dose in the presence of tissue heterogeneity must be standardized.
The advanced technologies offer exciting and potentially substantial advantages in radiation dose distributions, but given the current state of our knowledge, it can not be simply assumed that they help patients live longer or better. In fact, due to the reasons outlined above, we cannot even assume that the outcomes are as good as what is seen with “conventional” techniques in many cases! The possibilities of geographic miss, underdosage, or overdosage are real, even before taking into account the uncertainties in target delineation, deformation, motion, and heterogeneities.[ 3-8] Researchers should take into account the differential costs of replanning to account for tumor regression during treatment when calculating quality-adjusted life-years and cost-effectiveness.
Furthermore, two recent articles offer poignant reminders of how often the perception of academic clinicians that a new, experimental cancer treatment shall produce an outcome better than the standard treatment is proven wrong by prospective randomized trials.[10,11]
In part 2 of this article, which will appear in the next issue of ONCOLOGY, the authors describe the state of the science for various disease sites, with considerations of tumor control and toxicity rates after traditional conformal radiation therapy, and whether phase III trials support an additional benefit from the advanced technologies.
