Safety and efficacy of antiretroviral first-line regimens in adults
Date
2019
Authors
Venter, Willem Daniel Francois
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
As life expectancy in HIV-positive patients on antiretroviral therapy (ART) normalises, finding safer and cheaper ART options for decades of therapy, while preserving potency, is a healthcare priority. We surveyed doctors who were experienced in HIV care, using an anonymous online questionnaire, regarding World Health Organisation (WHO) HIV treatment guidelines, posing hypothetical scenarios, advising public health policy, treating family members or treating themselves (Chapter 2). Doctors approved of the WHO guidelines for public policy regarding antiretroviral drug choices (84%), CD4 initiation thresholds (54%) and monitoring (with the exception of isoniazid tuberculosis prevention (39%)). Doctors demonstrated concerns regarding the mitochondrial toxicity of stavudine (d4T) (lactic acidosis (38%) and lipoatrophy (37%)), as well as tenofovir (TDF) nephrotoxicity (55%) and efavirenz central nervous system side-effects (29%). They were significantly more likely to treat themselves, regardless of CD4 count, than public sector patients (95% odds ratio 3.33–7.33; P < 0.001).
We examined low-dose d4T (20mg twice daily) as a replacement for TDF in a large, non-inferiority, placebo-controlled, double-blind randomised study (Chapter 6). HIV-positive adults in South Africa, Uganda and India were randomised to either d4T or TDF (536 in each arm) with lamivudine and efavirenz for 96 weeks. Completion rates were 75.7% for d4T (n = 406) and 82.1% (n = 440) for TDF, with non-completion largely resulting from virological failure (6.2% d4T vs 5.4% for TDF; p = 0.60). For the primary endpoint (viral load <50 copies/mL), d4T was non-inferior to TDF (70.3%, 425/536 vs. 80.8% 433/536; difference = -1.49%, 95% CI = -6.3-+3.3; P < 0.01). Side-effects were higher with d4T, with 6.7% (n = 36) discontinuations as opposed to 1.1% for TDF (n = 6; P < 0.01). Lipoatrophy was more common with d4T (5.6%; 30) than TDF (0.2%, 1; P < 0.001). In summary, low-dose d4T showed non-inferior viral suppression but substantial toxicity.
The South African site participants were mostly recruited from ‘real-world’ primary health clinics, enabling analysis of the profile of patients entering public health programmes. Overall 771 patients were screened, who were mainly female (57%), with a mean age of 33.9 years (SD = 7.7), an initiation CD4 count of just over 200 cells/μL (median 208 cells/μl, IQR = 118–299 cells/ul) and high viral loads (39% > 100 000 copies/mL) (Chapter 7). Hepatitis B was highly prevalent (8%), 2% of patients had a haemoglobin<8g/dL, and neutropaenia was unusual. Creatinine clearance was >50mL/min in 99% of patients (microalbuminuria in 5%), raising questions about the need for routine pre-ART renal screening. CD4 cell count measurement is useful for predicting antiretroviral toxicity, immune recovery, immune inflammatory syndrome, prophylaxis indications and inform differential diagnoses. Using the same cohort, we analysed CD4 cell count variability between the referral, screening and baseline study visits (Chapter 8). All three CD4 cell counts were available for 553/771 patients. We found significant CD4 count variability across thresholds (350, 200 and 100 cells/μL), with 8% of patients referred below 350 cells/μL having screening values above that value (similar for baseline). Overall, 12% were misclassified at this threshold based on a CD4 cell count of > 350cells/μL on either screening or baseline (n = 64). A total of 17% of patients who had CD4 cell counts below 200 cells/μL at the time of referral had higher CD4 cell concentrations upon screening, and 24% were misclassified on either screening or baseline (n = 63). A total of 12–15% of participants were referred to the study when they had CD4 cell counts of under 100 cells/μL, but were found to be above 100 cells/μL upon screening or baseline, with 21% misclassified as having under 100 CD4 cells/μL on either one of these tests (n = 19). Based on the results of the referral, 17% of patients should have been classified as eligible for opportunistic infection prophylaxis, but they ineligible for opportunistic infection prophylaxis with subsequent testing. Around one in eight patients would be misclassified as falling within the 100 cells/μL threshold, with implications for cryptococcal antigen testing. An important recommendation was to alert clinicians to a range of possible (‘plausible’) values, so that misclassification does not occur through the application of rigid reported values.
In conclusion, doctors broadly agreed with WHO guidelines, although resistance to implementation of TB prophylaxis is of concern; low-dose d4T was efficacious regarding viral load suppression, but associated with unacceptable levels of mitochondrial toxicity; TDF was very safe in terms of renal function, but bone changes on DEXA were of concern; patients routinely entering care for ART had little in the way of background laboratory renal or haematological abnormalities, but high levels of hepatitis B; and CD4 counts should be interpreted carefully at critical thresholds that may trigger additional testing or initiation of opportunistic infection prophylaxis. Future research work focusing on addressing the toxicity of ART, while preserving virological efficacy, as well as rational evidence-based laboratory screening and interpretation at initiation of ART, is needed.
Description
Thesis submitted to the Faculty of Health Sciences, University of the Witwatersrand, in fulfilment of the requirements for the degree of Doctor of Philosophy.
Johannesburg, April 2019
Keywords
Citation
Venter, Willem Daniel Francois (2019) Safety and efficacy of antiretroviral first line regimens in adults, University of the Witwatersrand, Johannesburg, <http://hdl.handle.net/10539/28166>