A total of 89 HIV patients were enrolled for HIVDR study and 50 were males and 39 were females. Twenty-nine were successfully genotyped and among these 20 were males and 9 were females. The median (IQR) CD4 cell count was 569(443-760 cells/ml)and the median plasma viral Table 1 load was 11238.5(1000-744000copies/ml). Table 2

The table below summarizes the univariate analysis for the independent association of demographic variables with HIVDR mutations. No demographic variable was significantly associated with HIVDR mutations. Table 3

Of the 29 study participants who were successfully genotyped, 9 (31.03%) had a CD4+ count of 200–350 cells/μl, 8 (27.5%) had a CD4+ count of 351–500cells/μl and 12 (41.3%) had a CD4+ count of >500cells/μl.

The prevalence of NNRTI resistance was 25% (3/12) for participants with CD4+ cell counts of >500 cells/μl, 25% (2/8) for participants with CD4+ cell counts of 351–500cells/μl and 33.3% (3/9) for women with CD4+ cell counts of 200–350cells/μl.

Hundred and thirty-one (32.75%) out of 400 participants who were recruited into the mother study had VL≥ 1000 copies/ml, and we recruited 89 of them into our study. Of the 89 participants, 32.50% (29) had at least one drug resistance mutation. Among the 29 participants with HIVDR mutations, the prevalence of at least one HIVDR mutation to protease inhibitors (PIs), Nucleotide Reverse Transcriptase Inhibitors (NRTI), and Non- Nucleotide Reverse Transcriptase Inhibitors (NNRTI) were 6.47% ,46.76% and 46.76% respectively. Of the 29 participants who had HIVDR 19 (65.5%) had resistance to a drug they were currently taking and they needed to be switched to a better effective ART regimen. Among these 29 participants with available HIVDR results,18 were on 1st line treatment and 11 were on 2nd line ART. Fourteen of the 18 first line participants who experienced HIVDR needed to switch to 2nd line ,10 were on 3TC ,2 on abacavir and 2 on zidovudine based regimen. Of these (n=18) ,8 had the M184 mutation which confers resistance to 3TC and ABC, still it has also positive effects on the susceptibility of other NRTIs and can interrupt immunologic and clinical progression by reducing the viral fitness. It is also known to slow down the emergence of AZT, TDF and dd4T resistance.M184I/V mutations are linked with reduced viral multiplication both invivo and invitro. Normally M184I emerges before M184V due to the fact that it comes from a more common HIV-1 nucleotide substitution. Nonetheless, M184V surpasses M184I in several weeks of viral multiplication and this is common in majority of patients failing 3TC.Only 2 (4.22%) harboured the K65R mutation. Additionally, 5 participants on 1st line who needed ART switch had type 1 and 2 Thymidine analogue mutations (TAMs) combination (T215Y, L210W and D67N) and this combination is known to weaken the effectiveness of ABC and AZT. Type 1 TAMs have serious negative effect on virological response to ART regimen containing ABC or TDF and the might be the major reason why most of the participants on ABC containing regimen were failing.

All HIVDR mutations were categorized according to the WHO 2017 drug resistance list and Stanford HIV drug resistance database (SHIVDB). In 17.2% (5/29) of the individuals with HIVDR at least one major PI resistance mutation was identified, most commonly M46 and N88T had both 22.2% prevalence.

Overall, the 10 prevalent HIVDR mutations found in this study were M1844 (41.37%), K103 N (37.93%), G190A (27.58%), A98G (27.58%), K101E (24.13%), M41L (24.1%), V106M (20.68%), T215Y (17.24%), K219Q (13.79%) and M184MV (13.79%).

The prevalence of K103N and K101E was at 39.93% and 24.13% and these mutations are (individually or in concert) confer resistance to EFV and NVP (by 50 and 20-fold respectively). K101E normally occurs in combination with other NNRTI resistance mutations.

Among the NRTI mutations the most common TAM’s were M41L (24.1%), D67N (10.34%), T215Y (17.24%), and K70R (10.34%).

The ten most common mutations picked up in the NNRTI drug class included K103N (37.93%), V106M (20.68%), K101E (24.13%), A98G (27.58%). Figure1, Figure 2

M46I (6.89%) and N88T (6.89%) were the major HIVDR mutations and there were two PI accessory mutations L10I and L23F.Additional information on the distribution of mutations by drug class is shown in table 4.7 Figure 3

Our study findings provide valuable data on HIVDR among adolescents and children with virologic failure in Hurungwe, Zimbabwe. The study participants had similar characteristics i.e. those who were successfully genotyped and those who were not.

We observed HIVDR in 32.5% of the participants who had VL≥1000 having at least one mutation. More than half (62%) of those with HIVDR were resistant to at least one of their current ART drugs. This indicates that approximately 1/3 of adolescents and children who were failing ART require a drug switch and over 2/3 are experiencing ART failure as a result of other factors other than HIVDR for example non-adherence and sub-optimal dosing.

There is also a possibility that some of the participants with poor adherence could have HIVDR but with reversion to wildtype. The phenomenon of reversion HIV to wildtype in the absence of optimal drug pressure may complicate HIVDR result interpretation for the patients who experienced virologic failure but with no HIVDR.

Medium-high levels of HIVDR mutations are not unique to Hurungwe or Zimbabwe. A number of studies done in adults from Zimbabwe have reported a range of HIVDR from 10-48%. In one study conducted in children and adolescents in Zimbabwe, found that 31% had HIVDR, 29% cases were resistant to NNRTIs,3% were resistant to NRTIs. Above that, our findings indicated that most participants who had resistance to a current regimen were 1^st line failures and required a drug switch to the 2^nd line regimen. These findings highlight the potential implications of delayed clinical decision making in this special group. Chidamoyo Opportunistic Infection (OI) patients have no enough access to recommended testing onsite and they rely on Chinhoyi Central Hospital laboratory which is overwhelmed most of the times with samples from a number of sites in the provinces and sometimes it takes a month or more to get the result back. Inaccessibility of routine viral load tests and delays in turnaround times for VL results contribute to delayed decision making which might cause the emergence of HIVDR. Therefore, there is urgent need to strengthen mechanisms to ensure availability of VL results, improving tracking mechanisms for results and clinical supervision of healthcare workers to support timely decision making.

We also found that the prevalence of HIVDR mutation to 1^st line NNRTI and NRTI was the same (65%). Only 5 patients (17.24 %) were resistant to PI. The high levels of HIVDR to NNRTI supports the 2018 Zimbabwean treatment guidelines on the use of high genetic barrier drugs such as DTG for all age groups and PIs (Lopinavir-ritonavir) for specific groups including children who are not eligible for DTG as 1^st treatment in place of NNRTIs. However, numerous studies which were conducted in Zimbabwe and SSA region have revealed an increasing rate of primary NNRTI among adult patients. Increase in primary NNRTI resistance together with the increase of adolescents and children experiencing NNRTI resistance in this current study indicates an urgent need for optimizing HIVDR testing as a way of determining the best drug option, more importantly in those who were infected perinatally.

Our results support the need for routine HIVDR testing surveillance to determine prevalence of primary and acquired HIVDR in both ART-naive and experienced children and adolescents to guide therapeutic options as they develop into adulthood.

When it comes to HIVDR to NRTI, the prevalence of any TAMs was at 32.8% and the most common TAM s were common were M41L (24.1%), D67N (10.34%), T215Y (17.24%), and K70R (10.34%). TAM s confer resistance to zidovudine, stavudine and other NRTI. When compared to earlier studies from Zimbabwe, results from this current study shows a higher prevalence of TAM s among patients failing first line HIV treatment and with resistance.

Higher prevalence of TAMs (D67N and K70R) or TAMs in combination (T215Y with M41L) lessens susceptibility to Zidovudine (an important drug for most 2^nd line regimens). The TAMs of K70R/T/Q/N/E, K219Q/E, and D67N in combination effectively lowers the potency of zidovudine in a 2^nd line combination. The high prevalence of TAMs is a cause of concern although most of these participants (80.7%) have been on zidovudine or stavudine 1^stline ART at the time of enrolment into the study. Considering that these individuals were on abacavir or tenofovir, there is a chance that the TAMs observed emanated from TDR or to archived mutations from previous ART switches/exposures. Transmitted resistance and archived mutations have reported as potential causes of HIVDR. High NNRTI and NRTI resistance mutations may suggest the effect of previous policy driven first line ART switches over time that were done without any VL or HIVDR tests results. High prevalence of TAMs in our study brings to question the present-day practice especially in low income countries of switching patients failing first line regimens to second line regimens without HIVDR testing. With no HIVDR testing, we do not know what fraction of patients failing the first line ART where failing because of sub-optimal regimens or poor adherence. There is therefore an urgent need to provide HIV drug resistance testing at first line ART failure

Another major observation from our research was the high prevalence of the Y181C (10.3%) and Y188CE (3.44%) mutations, revealing resistance to 3^rd line therapy rilpivirine and etravirine, to which none of the participants had prior exposure to. Y181C and Y188C mutations has been demonstrated among 2^nd line failures earlier studies from SSA (5)(67). In our study, resistance to these drugs may be due in part to cross resistance with nevirapine and efavirenz which are widely used in PMTCT or first line ART. Regrettably, we were unable establish whether the HIVDR mutation to etravirine resulted from primary HIVDR transmission or developed as secondary HIVDR.

Findings from recent studies have showed the increasing outstanding role of integrase inhibitors over NNRTIs and PIs even in the presence of TAMs. These results may suggest that NRTIs are effective when used in together with integrase inhibitor dolutegravir (DTG)(68). Hence, the drive to introduce DTG as part of first and second line HIV care in Zimbabwe may be appropriate in the face of our evidence that third line ARV etravirine may potentially not be effective in approximately 3-5% of all youth failing on ART in Chidamoyo.

Gender differences in ARV treatment outcomes and drug-resistance mutations were also of interest in our study. Several studies from developed regions have not identified gender as a predictor for primary drug resistance. Differences between male and female participants were recorded in the levels of HIV-1 expressed mutations and HIVDR. In this study, we observed that the average number of HIV mutations was 4.9 in male and 4.5 in female participants respectively and this is in agreement with previous findings other studies, our sample size was low for us to conclude.

We also noticed that NNRTI-associated HIVDR was more common in individuals who had lower baseline CD4⁺ cell count.  Among the participants who had at least one HIVDR mutation, we noticed that any NNRTI resistance, the mutations G190A, K103N and K101E were associated with low CD4 cell (200-350cells/μl)(69). Though there is no scientific evidence from previous findings to support this observation, we suggest that this might be due to the fact that majority of our participants with HIVDR had NNRTI resistance.