Purpose: We systematically review the benefits and harms of intravesical therapies for nonmuscle invasive bladder cancer. Materials and Methods: Systematic literature searches were performed of Ovid MEDLINE (January 1990 through February 2016), the Cochrane databases and reference lists. Randomized and quasi-randomized trials of intravesical bacillus Calmette-Guerin, mitomycin C, gemcitabine, thiotepa, valrubicin, doxorubicin, epirubicin and interferon vs transurethral bladder tumor resection alone, and head-to-head trials of intravesical therapies were selected. Data were pooled using a random effects model. Results: Overall 39 trials evaluated adjuvant intravesical therapy vs transurethral bladder tumor resection alone. Bacillus Calmette-Guerin was associated with a decreased risk of bladder cancer recurrence (3 trials, RR 0.56, 95% CI 0.43-0.71) and progression (4 trials, RR 0.39, 95% CI 0.24-0.64) (strength of evidence low). Mitomycin C, doxorubicin, epirubicin and thiotepa were also associated with a decreased risk of recurrence, with no difference in risk of progression (strength of evidence low). There were 55 trials that compared one intravesical therapy agent against another. There were no differences between bacillus Calmette-Guerin vs mitomycin C in recurrence risk (RR 0.95, 95% CI 0.81-1.11), but bacillus Calmette-Guerin was associated with a decreased risk of recurrence in the subgroup of trials of maintenance regimens (RR 0.79, 95% CI 0.71-0.87, strength of evidence low). Bacillus Calmette-Guerin was associated with a lower recurrence risk vs doxorubicin, epirubicin, interferon alpha-2a, bacillus Calmette-Guerin plus interferon alpha-2b, and thiotepa (strength of evidence low to moderate). Bacillus Calmette-Guerin was associated with higher rates of local and systemic adverse events than other intravesical agents (strength of evidence low). Head-to-head trials showed no clear differences between standard and lower doses of bacillus Calmette-Guerin in recurrence, progression or mortality risk (strength of evidence low). Limited evidence suggested that bacillus Calmette-Guerin maintenance regimens are associated with reduced recurrence risk vs no further intravesical therapy in responders to induction therapy (strength of evidence low). Conclusions: For nonmuscle invasive bladder cancer several intravesical therapies are associated with a decreased risk of recurrence vs transurethral bladder tumor resection alone. Bacillus Calmette-Guerin is the only agent associated with a decreased progression risk vs transurethral bladder tumor resection alone, but may be associated with a higher risk of adverse events than other intravesical therapies, indicating trade-offs between potential benefits and harms.