Objective The aim of the study was to evaluate the efficacy of fosamprenavir/ritonavir (FPV/r) monotherapy in plasma and reservoirs in virologically suppressed patients. Methods A 48‐week, prospective, single‐arm pilot trial was carried out (trial registration: ISRCTN78584791). Patients receiving triple therapy FPV/r plus two nucleoside reverse transcriptase inhibitors (NRTIs) for at least the previous month, with viral load (VL) <40 HIV‐1 RNA copies/mL and no previous virological failure (VF) on protease inhibitors (PIs), were included in the trial and received FPV/r monotherapy (700/100 mg/12 h). VL and FPV/r levels by liquid chromatography‐tandem mass spectrometry (LC/MS/MS); limit of detection (LOD) 0.5 ng/mL in cerebrospinal fluid (CSF) were determined at week 24. VF was defined as VL >40 copies/mL in three consecutive samples or >500 copies/mL in two samples. Results Enrolment was prematurely stopped because of a high percentage of VF. Twenty patients (45% men; median age 43.5 years) were included in the trial. Nine patients (45%) presented therapeutic failure seven (35%) had VF, and two discontinued therapy. Resistance testing was available in five patients. One patient presented major PI mutations (54L, 32I and 47V) in addition to one minor mutation (13V), whereas two patients had minor PI mutations (10V+36I and 71T, respectively). The patient with major PI mutations switched from FPV/r to darunavir/r and VL was re‐suppressed. In the other six patients with VF, VL was re‐suppressed after the reintroduction of NRTIs. VL was <40 copies/mL in all CSF samples (n=10). Median amprenavir plasma levels were 2.5 μg/mL (range 0.7–8.6 μg/mL) at week 24 and 2.5 μg/mL (range 0.4–3.8 μg/mL) at VF. The CSF amprenavir concentration was 28.1 ng/mL (range 6.39–83.6 ng/mL), exceeding the reported 50% inhibitory concentration (IC50) range for CSF in nine of 11 patients. Conclusions The high percentage of patients with VF in our study suggests that the use of FPV/r in a simplification monotherapy strategy should be discouraged. Adequate amprenavir levels and undetectable VL in CSF were documented in all samples evaluated.