TPS4679 Background: The biology of metastatic RCC is extremely diverse, including a subpopulation of patients with indolent growth of metastases. Because of the acute and chronic toxicity and assumed non-curative nature of current systemic targeted therapy, a select subset of patients may be better served with initial surveillance. Such an approach may allow for deferral of systemic therapy to minimize overall treatment-related toxicity burden while maintaining treatment benefit. Methods: A prospective phase II trial is being conducted to characterize the clinical course of patients with mRCC who defer initial systemic treatment. Appropriate patients are selected by the treating physician based on an observed indolent growth pattern. As such, there is a 12 month window allowed between the first diagnosis of metastatic disease and study entry. Patients must be treatment-naïve, asymptomatic, with histologically confirmed mRCC and clinically-evident, measurable disease to be eligible. Radiographic assessment is performed at baseline, every 3 months for year 1, every 4 months for year 2, then every 6 months. The primary objective is to characterize the clinical outcome of patients on observation in terms of time to RECIST defined disease progression and time to initiation of systemic treatment. Secondary endpoints include measurement of disease-related symptoms and depression/anxiety using standardized questionnaires (FKSI-DRS and HADS), as well as correlative endpoints analyzing the immune response over time for which blood is being collected for immune assays (TH1/TH2 phenotype, Tregs). Pts remain on study until initiation of systemic therapy due to radiographic disease progression, development of disease-related symptoms, withdraw of consent or clinical change that renders the patient unacceptable for further observation. Local therapy (e.g. surgery, radiation) during the observation period is permitted. Currently, 29 patients have been accrued at 5 collaborating sites. The target accrual of 50 pts provides adequate power for the primary descriptive endpoint as well as 80% power to detect changes from baseline for the correlate endpoints based on a two-sided Wilcoxon signed rank test.