Treatment failure and drug resistance not associated with rapid death or illness in Côte d’Ivoire study

HIV-positive individuals who experienced failure of their first three-drug antiretroviral regimen and who developed drug resistance did not have a higher risk of serious disease or death during a 20-month follow-up according to the findings of a longitudinal study published in the May 31^st edition of AIDS.

The study also found that the only two factors significantly associated with a risk of immunological failure - a decline in the CD4 cell count below 200 cells/mm³ - were a CD4 cell count below 200 cells/mm³ at baseline and the presence of a major resistance mutation on entry to the study.

A large number of AIDS patients in sub-Saharan Africa are now on antiretroviral therapy (ART) as a result of concerted national and international efforts. However, with increasing access to antiretrovirals in Africa, an important public health concern has emerged: the spectre of antiretroviral drug resistance.

Several studies have demonstrated that the rate of primary resistance to antiretroviral drugs is low in sub-Saharan Africa but that there is rapid emergence of drug resistance when treatment fails to suppress viral load, due in part to the nature of the drugs used in resource-limited settings. Resistance to lamivudine and to the non-nucleoside reverse transcriptase component of triple drug therapy emerges at high frequency in developed world and developing world cohorts when first-line therapy fails.

There is a lack of information about the patterns of primary and secondary drug resistance in HIV isolates currently in circulation in African countries. This evidence is required to formulate treatment guidelines regarding the optimal first and second-line antiretroviral regimens for countries where access to HIV therapy is being scaled up.

A team of French and Ivoirian investigators have addressed this problem in a longitudinal study of adult AIDS patients on ART in Côte d’Ivoire. The patients had taken ART for a median of three years. The team studied the association between the presence of resistance mutations and treatment outcomes.

Prior to the study, a cohort of 723 HIV-infected adults were followed up in the ANRS 1203 study in Abidjan from 1996 to 2003. When ANRS 1203 ended, ACONDA/ISPED, a new programme of HIV care and treatment, was launched. Participants in the reported study were all HIV-infected adults who received ART and were followed up under ANRS 1203 and subsequently enrolled in ACONDA/ISPED. Follow up included monthly clinical visits and six-monthly CD4 cell counts.

Laboratory investigations included CD4 cell counts, HIV viral load tests, and for patients with detectable viral loads, genotypic resistance testing at baseline. A patient’s viral mutations were classified as major or minor. A strain was considered drug resistant if it displayed major drug resistance mutations for any of the three major drug classes.

At study entry, the median previous time on potent antiretroviral therapy was 37.4 months, the median nadir CD4 cell count was 122 cells/mm³; 62 patients (58.5%) had undetectable viral loads, 21 patients (19.8 %) had detectable viral loads with no major resistance mutations, while 23 patients (21.7 %) with detectable viral loads had one or more major primary mutations.

For the 23 patients with major mutations, 16 presented major mutations for one class of drugs and seven had mutations for two drug classes. A low CD4 count (less than 200 cells/mm³) was significantly associated with at least one major mutation (p = 0.01).

During follow-up, only ten of the 23 patients with major resistance mutations were switched to a new ART regimen to which the patient had no genotypic resistance. The authors say that one of the major reasons for a lack of switching was the severe political unrest affecting Côte d’Ivoire during the period of the study.

Of the remaining 13 patients, three were still on their first-line regimen and ten were already receiving a second regimen at entry.

During follow-up one patient died and one was lost to follow-up; both had major resistance mutations. Twenty-nine patients (about one third of whom had major resistance mutations) experienced 43 new episodes of serious morbidity. Immunological failure was defined as a CD4 cell count below 200 cells/mm³. Baseline mutations at the beginning of the study were not associated with the risk of serious morbidity at follow-up.

There was no significant difference in mortality and morbidity between patients with virologic failure and major mutations, and patients with undetectable viral loads. The 18-month probability of remaining alive and without severe morbidity was 0.79, 0.86, and 0.69 in patients with undetectable viral loads, in patients with detectable viral loads without major resistance mutations (p = 0.91), and in those with detectable viral loads with major resistance mutations (p = 0.19), respectively.

CD4 counts remained stable for the majority of patients with virologic failure at baseline. The median gain in CD4 cell count between study entry and the date of the last available CD4 count was + 129 cells/mm³ in patients with undetectable viral loads, +51 cells/mm³ in those with detectable viral loads with no mutations and +3 cells/mm³ in those with detectable viral loads with resistance mutations.

A multivariate analysis demonstrated that CD4 counts less than 200 cells/mm³ and the presence of resistance mutations at entry were significantly associated with immunological failure.

“On the one hand,” write the authors, “the 20-month clinical and immunological outcomes of patients with resistance at baseline were clearly compromised compared with patients with no resistance. On the other hand, these patients had reasonable viral load values at baseline. During the 20 month follow-up, their CD4 cell counts remained stable and close to 200 cells/mm³. Although they tended to have higher morbidity rates, most were curable diseases and only one patient died.”

“In other words, their medium-term outcomes were impaired compared with patients without resistance mutations,, but their antiretroviral treatment still protected them from immunological breakdown. The main reason for this is probably that they continued to receive at least one or two drugs against which their viruses had no resistance.”

The findings represent the first published study on the impact of HIV-1 genotypic drug resistance mutations on treatment outcome in adult African AIDS patients. The study, as pointed out by the authors, has several limitations.

These include the fact that the study participants started ART while on another cohort study and resistance mutations in those who died remain unknown, patients were on different ART regimens and had different histories of regimen modification, and no adherence evaluation was done.

An accompanying editorial points out the need to interpret the data with caution since the Côte d’Ivoirian cohort might not be representative of all resource-poor settings. The authors warn that long-term failing treatment with nucleoside analogues may lead to the development of thymidine analogue mutations, leading to cross-resistance with second-line nucleosides like abacavir and didanosine.

They also express concern about the potential for transmission of drug-resistant viruses to sexual partners if individuals are maintained on failing regimens for long periods.

They suggest that tenofovir-based regimens might be preferable in first-line treatment in order to preserve AZT for second-line use, and that NRTI-only regimens (such as the AZT/3TC/tenofovir regimen being tested in the DART study) might limit the extent of cross-resistance.

They also say that research into alternative, cheaper and more practical methods of viral load testing “is currently half-hearted, and needs to be accelerated. Virological monitoring should be made more available and less expensive in resource-limited settings.”