Lopinavir, an HIV-1 protease inhibitor, was administered in combination with ritonavir to increase drug bioavailability of lopinavir [15]. LPV/r has been shown to be effective in patients infected with SARS-CoV [16]. LPV/r also improved clinical parameters in MERS-CoV-infected marmosets and mice [17, 18]. Our study has, for the first time, shown that the lack of LPV/r treatment was independently associated with prolonged SARS-CoV-2 RNA shedding in non-critically ill patients with COVID-19. Administration of LPV/r treatment could shorten the duration of viral shedding compared with no LPV/r treatment. Interestingly, subgroup analysis revealed that the early administration of LPV/r treatment within 10 days from symptom onset, but not late administration (initiated after 10 days), could shorten the duration of SARS-CoV-2 RNA shedding compared with no LPV/r treatment. A recent randomised controlled trial showed that LPV/r treatment could not provide additional benefit, including in relation to viral shedding, beyond standard-of-care in hospitalised severely ill adult patients with COVID-19 who required a range of ventilatory support modes [9]. However, a post hoc subgroup analysis of the LPV/r treatment in the randomised controlled trial found that earlier administration of antiviral treatment could accelerate the clinical recovery and reduce the mortality, which was consistent with our findings. Admittedly, patients recruited in the trial were in an advanced phase of infection (as evidenced by the 25% mortality in the control group). In contrast, most patients (99.2%) in our cohort were non-critically ill patients with COVID-19, due to triage strategies. Because LPV/r was not particularly designed to act against SARS-CoV-2, it might reasonably be anticipated that LPV/r action on SARS-CoV-2 was not as effective as neuraminidase inhibitor action on influenza [19]. The median duration of viral shedding, even in patients who received LPV/r treatment within 10 days from symptom onset, was 19 days in the current study, demonstrating that LPV/r treatment could not completely inhibit viral replication. In this regard, earlier administration of LPV/r therapy after symptom onset, to patients who were not in the late phase of infection, may be crucial to determine the efficacy. Nevertheless, randomised clinical trials remain crucial to test whether earlier administration of LPV/r treatment could shorten the duration of viral shedding, accelerate the clinical recovery and improve survival in patients with COVID-19, especially in non-critically ill patients.