Cannabis use is associated with a lower likelihood of presence of HIV drug resistance mutations in a retrospective cohort of adults with HIV

Introduction

Substantial strides have been made to turn HIV, a disease in which there is no vaccination or cure, into a chronic but manageable disease with the use of anti-retroviral therapy (ART) [1]. The goal of ART is to achieve viral suppression through significantly reducing viral replication, and thus, reduce the neurocognitive effects that accompany chronic HIV infection. One significant barrier to combating HIV is the emergence of anti-retroviral drug resistance mutations (ARV-DRM) within the HIV-infected population. HIV is particularly susceptible to replication issues inherent within the action of the virus’s reverse transcriptase, which are a source of ARV-DRM [2], 3]. Although these drug-resistant variants of HIV are less fit, selective drug pressure allows for these variants to continue to replicate within the infected individual and reduces the efficacy of ART-mediated viral suppression. One of the predominate issues with the development of ARV-DRM and the subsequent lack of viral suppression is cerebrospinal fluid (CSF) escape [4]. CSF escape occurs in 4–20 % of HIV-infected adults and is associated with an increase in neurocognitive symptomology and faster progression of the HIV infection, including HIV-associated dementia [4], 5]. CSF escape is diagnosed by meeting one of two criteria: either by having higher HIV RNA levels in the CSF than within the plasma, or by having differences in quantity or type of ARV-DRM in the CSF HIV RNA compared to plasma HIV RNA [4], 6], 7]. Not only does this pose a risk for individuals already infected with HIV, but also allows for transmitted drug resistance (TDR) to another individual.

ART has been adjusted to be used as a prophylaxis for individuals who are currently HIV-negative but at high risk for HIV infection. Pre-exposure Prophylaxis (PrEP) prevents viral replication and thus prevents the establishment of a chronic infection [8]. The efficacy of PrEP has been reported in a number of trials and although highly dependent on the level of medication adherence, has shown significant reduction in HIV infections among high risk individuals [9]. In humans, there have only been six reported cases of HIV infection in individuals who were taking PrEP and adhering to the medication regimen [10]. Of those six cases, five of them were infected with an HIV strain that had ARV-DRM. Although the number of individuals utilizing PrEP is growing, the presence of ARV-DRM within the HIV-infected population undermines these efforts [11], 12]. While PrEP has been an effective prophylaxis treatment, there has been a growing interest in research to understand the development of ARV-DRM in the PrEP era and whether there are specific precipitating factors that increase the incidence of ARV-DRM.

Previous research has identified that drugs of use, misuse and addiction affect every aspect of HIV, including viral replication, transmission, pathogenesis, treatment efficacy and adherence [13], 14]. Stimulants (such as cocaine, crack cocaine, methamphetamine, etc.) have a well-documented association with adverse HIV outcomes (e.g. low CD4 counts, viral load) both in clinical and basic science research [15], [16], [17], [18], [19], [20]. Drug users are more likely to experience gaps in treatment adherence [21], 22] and drugs of use, misuse and addiction may impact ART metabolism and distribution in the brain [23], [24], [25]. However, most studies on the effect of addictive substances on viral suppression do not specifically examine whether this reduced viral suppression places an HIV positive individual at an increased risk for ARV-DRM. The current study sought to investigate whether there was a relationship between a history of substance use and the likelihood of having HIV ARV-DRM among a clinical sample of HIV positive individuals. Previous findings have suggested that substance use may be a mechanism that leads to drug resistance mutations among the HIV positive population. Thus, we hypothesized that there will be a significant and positive relationship between HIV ARV-DRM and substance use in this study cohort.

Materials and methods

Results

Discussion

In this retrospective study of HIV positive participants, cannabis use was significantly associated with reduced odds of having ARV-DRM in a subsample of participants that did not test positive for polysubstance use. This relationship is also supported by a significant result in a model that includes the full sample and adjusts for the interaction between Cannabis and polysubstance use and is potentially related to our finding that Cannabis use is associated with lower odds of high-viremia. Taken together, this provides evidence for a negative relationship between cannabis use and presence of ARV-DRM, and no significant positive relationships between substance use and presence of ARV-DRM.

The significance of cannabis use in relation to ARV-DRM only emerges when the interaction between cannabis and polysubstance use is included in the analysis, despite the interaction term itself not being statistically significant. This pattern suggests a lack of stability in these relationships, likely due to the small sample size, the complexity and overlap of substance use behaviors, and the rarity of ARV-DRM [33]. A larger sample size may help clarify these associations and provide more robust evidence. A sensitivity analysis demonstrated that when the rare class of polysubstance use with cannabis and ARV-DRM was increased from n=1 to n=2 through simulating a representative additional sample of 42 participants, Cannabis use was marginally significant (p=0.06, OR=0.3).

Of particular interest in this study were the results of Cannabis being a potentially protective factor for the development of an ARV-DRM. Cannabis is widely recognized for its actions on the cannabinoid type 1 and 2 (CB1 and CB2) receptors. CB1 is typically considered psychotropic while CB2 receptors are considered the anti-inflammatory cannabinoid receptor [34]. Several studies have found that CB2 agonists or THC reduces blood-brain barrier dysfunction, activation of macrophages, and HIV/SIV replication [35], [36], [37]. Our analysis found a significant negative relationship between cannabis and high viral load. Thus, a possible explanation for the reduced likelihood of HIV ARV-DRM in cannabis using participants could be that with reduced neuroinflammation and attenuation of viral replication, there is a reduction in the accumulation of viral quasispecies that acquire an advantageous drug resistance mutation. This notion is supported by a recent study that found reduced odds of tissue harboring HIV-1 DNA and viral DNA copies in brain and peripheral tissue in people using cannabis [38]. In that study, inflammatory cytokine levels were also significantly lower in lymphoid tissues [38]. Others studies have also found reduced inflammatory markers in the CSF and blood in recent cannabis users [39].

We also performed an analysis to determine whether there were relationships between drug use, CSF escape and ARV-DRM. No significant relationships were found between cannabis and CSF escape, or between ARV-DRM and CSF escape. In light of these results, future research should further investigate the effects of cannabis on the development of HIV ARV-DRM and on CSF escape.

Although there is some evidence of the potential benefits of cannabis use with HIV infection, there are also studies that do not support this conclusion. Roth [40] and Thames [41] have shown that cannabis over-suppresses the immune system which can allow for increased viral replication and worsened neurocognitive symptomology [40], 41]. In addition to these studies on potential negative effects of cannabis, several studies have found no differences in the effect of cannabis in viral load [42], 43] in contrast to our cohort.

Based on prior studies [44], [45], [46], [47], [48], [49] we expected to see an increased risk for ARV-DRM among stimulant using participants and intravenous substance using participants. One of the reasons this study may not have found similar results could be the small percentage of participants who endorsed solely stimulant use and intravenous substance use. Moreover, the participants that used stimulants and used substance intravenously also used cannabis, which could have overshadowed the harmful effects of these behaviors. Although we did not find support for increased ARV-DRM in stimulant or intravenous substance users within our study, the positive significant relationship between amphetamines and CSF escape suggests that these relationships may be complex and worth further study.

The current study has several limitations that should be addressed. First, the proportion of the sample that had HIV strains with ARV-DRM was very low, only accounting for slightly over 3 % of the total sample. Due to this low incidence, it is difficult to make substantial claims about the results our data shows. We performed an analysis of data from a prior study [4] that is comprised of a cohort of participants from the CHARTER-HNRC and CHARTER-NNTC cohorts. ART adherence, which can contribute to the development of ARV-DRMs, was not evaluated. Drug users have been reported to have reduced ART adherence [50]. However, several studies note that cannabis users did not have issues with ART adherence [22]. However, one study found that individuals with cannabis dependence [22] and older individuals has suboptimal ART adherence [21]. Our analysis which indicated that cannabis users had reduced risk of ARV-DRMs does not suggest that these individuals had suboptimal ART adherence. If these cannabis users did have suboptimal ART adherence, it may suggest a larger effect than we have measured. The prior analysis of the cohort [4] found that PI regimens increased the risk for mutations. While we could not perform an analysis to examine whether individuals on PIs were more or less likely to have used cannabis, we do not feel that cannabis use would impact the probability of an individual to be on a PI regimen.

With the rise of the use of PrEP, there is a need to monitor changes in HIV quasispecies in order to prevent spread through resistance mutations. However, it’s important to note that new PrEP formulations are on the horizon and some regimens are more resilient against ARV-DRMs. Consequently, the risk of transmission of HIV with ARV-DRMs may decrease as therapeutics evolve.

Future research should have a sample with a larger proportion of HIV ARV-DRM or equal representation of ARV-DRM patients and non-ARV-DRM patients. Additionally, the way in which substance use was analyzed has notable limitations. Within the original data collection there was an inconsistent number of participants visits for urinalysis across the sample. To account for this, a participant was coded as a substance user with a single positive urinalysis screening. Subsequent studies with a higher frequency of ARV-DRM participants should use ranked analyses under a consistent number of urinalysis screenings or use a psychiatric assessment of substance use disorders. Additionally, drug use was evaluated based on urinalysis data which does not track prior medical diagnoses (substance use disorders, dependence etc.) or self-reported drug use.

In summary, this study provides preliminary data for the association between substance use and HIV ARV-DRM. Specifically, this research found that cannabis use is significantly associated with a decreased likelihood of HIV ARV-DRM. More research should focus on the potential protective and risk factors of addictive substances in clinical and in vivo settings to identify the potential mechanism of action that these substances are having on HIV pathology, replication, and neurocognitive symptomology. Finally, in the era of PrEP, it is becoming urgently necessary to understand any factors that may affect the development of ARV-DRM and the spread of these quasispecies through the population.

Conclusions

The presence of HIV transmitted drug resistance is of growing concern within the clinical community during the era of pre-exposure prophylaxis for individuals at high risk for HIV infection. Infection by an HIV quasispecies that is resistant to the standard prophylaxis medication renders the treatment ineffective and allows for the establishment of chronic HIV infection. To date, there is little published work that has investigated the association between drugs of use, misuse and addiction and the risk of development of anti-retroviral drug resistance mutations within a clinical sample of HIV positive individuals. We have found that cannabis use is significantly associated with a decreased likelihood of HIV ARV-DRM and suggest that cannabis use may be a protective factor against the development of anti-retroviral drug resistance mutations. More research is necessary to identify the mechanism in which addictive substances act on HIV viral suppression and on the development of anti-retroviral drug resistance mutations.