Let's Get SEERious: More Accurate Staging With Consequent High Management Impact Is Not Just Stage Migration

Abstract

issue in heath care resourcing. Although individuals with non–smallcell lung carcinoma (NSCLC), their physicians, and third-party health care payers deserve reassurance that the increasing use of positron emission tomography (PET) is beneficial, we have significant concerns regarding the methodology applied and, consequently, the validity of the resulting conclusions. Most importantly, we feel that all-cause survival is an inappropriate primary end point for judging the benefits of a diagnostic test. This parameter ignores the potential of a superior investigation to guide more efficient and appropriate use of therapeutic interventions and to induce changes in management that may influence quality of life or progressionfree survival in particular patient subgroups without necessarily affecting long-term survival. This is particularly relevant in a disease like NSCLC in which current therapeutic interventions have suboptimal efficacy and are associated with substantial cost and toxicity. Nevertheless, because it has been previously established that the dominant effect of applying PET to NSCLC staging is the detection of unrecognized metastases, 2,3 a logical assumption must be that palliative therapies would be more frequently applied with increasing use of PET. If PET inappropriately upstages patients and results in patients’ denial of potentially curative therapies, overall survival should logically deteriorate. The authors’ finding that all-cause survival did not change significantly with the increasing use of PET supports a conclusion that PET may reduce morbidity associated with futile therapies without negatively affecting overall patient outcomes. We believe that if any conclusion can be reached, their data thus support rather than repudiate the utility of PET in NSCLC. As there is currently no curative therapy available for stage 4 NSCLC, it is unreasonable to expect a large change in overall survival with the application of PET. The earliest study end point of 2-year overall survival used by the authors is, we contend, too long to reliably detect improvements in survival associated with better selection and delivery of palliative treatments in such patients. In our own recently published prospective trial of PET/computed tomography in NSCLC, the median survival of patients with stage 4 disease on PET was less than a year and almost all patients in this stage group died within 2 years. 4 Moreover, patients who died as an immediate complication of futile therapy would also not be detected because the study excluded patients who died within 2 months.