Spanish study reports associations between COVID-19 and HIV treatment including NRTIs

A large observational Spanish study looking at the incidence and severity of COVID-19 in HIV positive people on ART diagnosed with COVID-19 has reported a potential benefit from TDF/FTC. [1]

However, as with all observational data, these results, published in Annals of Internal Medicine, need to be interpreted with caution as they haven’t been seen in other cohort studies. Most importantly, the study doesn’t appear to report or adjust for comorbidities, which are critical for COVID-19 outcomes.

This national cohort included 77,590 HIV positive adults at 60 HIV clinics in Spain whose records were included from 1 February to 15 April 2020.

Of these, 236 people were diagnosed with PCR-confirmed COVID-19, 151 were hospitalised and 15 were admitted to the ICU. There were 20 deaths: 5 in ICU, 12 in hospital (but not in ICU) and 3 who were not hospitalised.

The risk per 10,000 for HIV-positive vs general population (standardised for age and sex and excluding health workers) was 30.0 vs 33.0 for COVID-19 diagnosis and 3.7 vs 2.1 for death.

The analysis of ART use focused on the NRTI component due to modelling data supporting potential activity against SARS-CoV-2 from tenofovir disoproxil (TD), tenofovir alafenamide (TAF), abacavir (ABC), emtricitabine (FTC) and lamivudine (3TC).

The main results reported lower risk for diagnosis and hospitalisation associated with ART containing TDF/FTC, with no related ICU admissions or deaths. However, the multivariate models adjusting for ART don’t appear to have adjusted for other clinical factors associated with COVID-19 outcomes. See Table 1.

Table 1: COVID-19 outcomes by NRTI (note: not adjusted for COVID-19 factors)

Note: Table includes 186 people on 3-drug ART. 50 participants were on 2-drug ART.

comment

We reported this study because of the high profile generated from the top-line conclusion. However, it is surprising the paper was published given the important cautions, where associations could easily be from confounding factors related to COVID-19.

Although this is a larger cohort, including one of the largest groups of people with HIV/COVID-19, the numbers are small when looking at the impact of ART components.

Also, while the paper reports on HIV treatment, it includes no information are reported on other risk factors for COVID-19 (other than age and gender) including prevalence and severity of comorbidities. The lack of information on kidney disease at baseline and as an outcome is especially important given the focus on tenofovir and TAF.

It is also not clear from the paper whether this was a prospective or retrospective study.

NRTI component	n	Risk for C-19 diagnosis	Risk for hospitalisation
TAF/FTC	100	39.1 (95%CI: 31.8 to 47.6)	20.3 (95%CI: 15.2 to 26.7)
TDF/FTC	21	16.9 (95%CI: 10.5 to 25.9)	10.5 (95%CI: 5.6 to 17.9)
ABC/3TC	57	28.3 (95%CI: 21.5 to 36.7)	23.4 (95%CI: 17.2 to 31.1)
Other	8	29.7 (95%CI: 22.6 to 38.4)	20.0 (95%CI: 14.2 to 27.3)