Cancer patients are often told that they have """"""""no evidence of disease"""""""" after initial therapy. Yet, it is crucial for these """"""""cured"""""""" patients (pts) to have careful follow-up and monitoring for recurrence. Unfortunately, by the time such recurrence is detected by scan or symptoms, it is often too late for curative intervention. Here we propose to use molecular markers to detect minimal residual disease (MRD) immediately after completion of therapy. We use as our model metastatic neuroblastoma (NB), an orphan childhood cancer that is difficult to cure because of its tendency to relapse after near complete remission. We propose to target subclinical NB as an alternative to the traditional approach of waiting for signs or symptoms of gross disease. Doing so requires an accurate quantification of MRD in order to: (1) detect recurrence much earlier than the conventional clinicopathological methods and (2) use early response indicators for timely treatment interventions. Development of MRD measurement will also facilitate (3) assessment of the efficacy of novel therapies. Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) is a highly sensitive method for measuring transcripts from viable tumor cells circulating in bone marrow (BM), peripheral blood (PB), and stem cell harvests. We have demonstrated that GD2 synthase and tyrosine hydroxylase (TH) are useful MRD markers. Nevertheless, given the heterogeneity among and within NB, studying multiple markers will enhance detection. Using tumor expression-arrays, we discovered cyclin D1 (CCND1) as a highly promising MRD marker. We propose to test the clinical utility of a core marker set (GD2synthase,TH, CCND1) for measuring MRD in 136 pts undergoing adjuvant therapy by using archived samples in the R21 phase, and prospectively collected fresh samples in 129 pts for the R33 phase. We will also complete the analysis and validation of the remaining 10 novel markers discovered by expression array analysis. Besides proving their utility as accurate measures of MRD, this application tests the hypothesis of using early MRD response for prognostication, potentially providing a guide for choosing therapies. We also aim to accelerate technology transfer to the clinic by validating one-step BM and PB sampling methods. Armed with the tools to collect samples expeditiously and to measure MRD accurately, the paradigm of treating subclinical NB can then be tested in multicenter randomized studies. ? ?
