Changing Treatment Early Enough to Matter
This issue we are discussing several important topics in molecular imaging and medicine. One addresses the need for developing tools not only to monitor therapeutic responses, but also to stratify patients into the treatments that are most likely to result in significant responses. This concept can be applied not only to oncology, but also is highly relevant in patients with neurologic and cardiac diseases.
In oncology, treatment response assessments are traditionally performed late in the treatment process, often when therapeutic approaches can no longer be adjusted. These tests then serve as prognosticator—but have limited impact on patient outcome. PET/CT imaging mostly with 18F-FDG can, however, be employed to determine therapeutic responses early after the start of therapy. This allows treating physicians to adapt their therapeutic strategies to responses as determined by glucose metabolic activity of tumors (risk adaptive therapy). Several approaches for providing objective response assessments have been proposed and these are increasingly used in the clinic. For instance, Wahl et al recently introduced PERCIST (PET Response Criteria in Solid Tumors) criteria as a novel way to standardize PET-based treatment response assessments (J Nucl Med 2009).
These and other approaches now provide, for the first time, guidelines that can be used by imaging experts and treating physicians to arrive at more rational treatment decisions.
Response rates to a variety of therapeutic interventions have remained disappointingly low. In other words, many cancer treatments fail to induce lasting responses. There is, therefore, a need to develop imaging probes that would permit predictions of therapeutic responses before therapies are started or at least early in the treatment course. This can be accomplished by using PET imaging for phenotyping of tumors. Probes of tumor metabolism (glucose, amino-acid, protein, lipid and nucleotide) as well as imaging markers of angiogenesis, hypoxia, gene expression and many more have been developed. If deployed appropriately, these imaging probes could help to better stratify patients into likely responders vs. non-responders.
In some instances, the combination of several imaging probes might theoretically be beneficial for more comprehensive tumor phenotyping. This approach would obviously increase the radiation exposure to patients. Given the current public awareness and concerns regarding radiation dose, a discussion about real and perceived cancer risks associated with molecular imaging is necessary. To my knowledge, there is little epidemiologic data that associate medical imaging with increased cancer risk. If used appropriately, the risk/benefit ratio of medical imaging is extremely low. The real risk is that superficial headline-oriented discussions result in patients avoiding truly necessary diagnostic imaging tests. As you’ll see, this issue addresses this topic in a way that hopefully provides some reason in this emotionally charged debate.