Background: In clinical trials, toxicity leading to tenofovir disoproxil fumarate (TDF) discontinuation is rare (3% by 2 years); however in clinical practice it seems to be higher, particularlywhen TDF is co-administered with ritonavir-boosted protease inhibitors (PI/r). Aims of this study were to assess the rate of TDF discontinuations in clinical practice and to identify factors associated with the risk of stopping TDF. Methods: All antiretroviral treatment (ART)-naive patients initiating a TDF-based regimen were selected from the ICONA Foundation Study cohort. The primary outcome was TDF discontinuation regardless of the reason; secondary outcome measures were TDF discontinuation due to toxicity and selective TDF discontinuation (that is, TDF discontinuation or substitution, maintaining unchanged the remaining antiretroviral treatment). Results: 3,618 ART-naïve patients were included: 54% started a PI/r-based and 46% a NNRTIbased based regimen. Two-hundred-seventy-seven patients discontinued TDF and reintroduced ART within 30 days without TDF. The probability of TDF discontinuation regardless of the reason was of 7.4% (95%CI:6.4-8.5) by 2 years and 14.1% (95%CI:12.2-16.1) by 5 years. The 5-year KM estimates in the PI/r vs. NNRTI group were 20.4% vs. 7.6%, respectively (log-rank p = 0.0001), for the outcome of stopping regardless of the reason, and 10.7% vs. 4.7% (p = 0.0001) for discontinuation due to toxicity. PI/r use and lower eGFR were associated with an increased risk of discontinuing TDF. Conclusion: In our cohort, the frequency of TDF discontinuations was higher than that observed in clinical trials. Co-administration of TDF with PI/r was associated with an increased rate of TDF discontinuations. Further studies are needed to clarify the mechanisms that might have led to this outcome.