The main aim of this paper was to compare results of Genescan and real-time PCR methods in order to detect contamination in harvests from patients with follicular lymphoma. The secondary goal was to evaluate the efficacy of Rituximab as an in vivo purging agent. A total of 23 patients had been treated with CHOP followed by either high-dose therapy (12 patients) or high-dose plus Rituximab (11 patients), both followed by autologous transplantation. Results show that 86% of harvests from patients treated with Rituximab were PCR-negative compared to 14.3% from controls. Real-time PCR was more sensitive than Genescan PCR; quantitative analysis revealed a correlation between the amount of contamination in the harvests and relapse after transplantation. Whereas all patients reinfused with negative aphereses achieved complete remission and showed a significantly better 5-year PFS (100%) compared to those reinfused with contaminated samples (41%), a very low amount of contamination does not appear to negatively affect outcome, suggesting that determination of a cutoff in the contamination level of harvests could be useful. Results suggest that real-time PCR is superior to Genescan PCR to select transplantable harvests and confirm the ability of Rituximab as an in vivo purging tool for follicular lymphoma.