Hepatitis B testing and treatment in HIV patients in The Gambia—Compliance with international guidelines and clinical outcomes

Background Compliance with WHO guidelines on HBV screening and treatment in HIV-coinfected patients is often challenging in resource limited countries and has been poorly assessed in sub-Saharan Africa. Methods Between 2015 and 2016, we assessed physician’s compliance with WHO guidelines on HIV-HBV coinfection in the largest HIV clinic in The Gambia, and the hepatic outcomes in HIV-HBV coinfected patients as compared to randomly selected HIV-monoinfected controls. Results 870 HIV-infected patients regularly seen in this clinic agreed to participate in our study. Only 187 (21.5%, 95% CI 18.8–24.3) had previously been screened for HBsAg, 23 (12.3%, 95% CI 8.0–17.9) were positive of whom none had liver assessment and only 6 (26.1%) had received Tenofovir. Our HBV testing intervention was accepted by all participants and found 94/870 (10.8%, 95% CI 8.8–13.1) positive, 78 of whom underwent full liver assessment along with 40 HBsAg-negative controls. At the time of liver assessment, 61/78 (78.2%) HIV-HBV coinfected patients received ART with 7 (11.5%) on Tenofovir and 54 (88.5%) on Lamivudine alone. HIV-HBV coinfected patients had higher APRI score compared to controls (0.58 vs 0.42, p = 0.002). HBV DNA was detectable in 52/53 (98.1%) coinfected patients with 14/53 (26.4%) having HBV DNA >20,000 IU/L. 10/12 (83.3%) had at least one detectable 3TC-associated HBV resistance, which tended to be associated with increase in liver fibrosis after adjusting for age and sex (p = 0.05). Conclusions Compliance with HBV testing and treatment guidelines is poor in this Gambian HIV programme putting coinfected patients at risk of liver complications. However, the excellent uptake of HBV screening and linkage to care in our study suggests feasible improvements.

Introduction Chronic hepatitis B virus (CHB) infection is a major public health problem and a leading cause of morbidity and mortality globally. It affects approximately 250 million persons worldwide [1] and accounts for 650,000 deaths annually [2]. Without effective interventions and treatment, CHB infection will lead to an estimated 11.8 million deaths by 2030, primarily as a result of cirrhosis and hepatocellular carcinoma (HCC) [3]. Most of these deaths will occur in low-income countries (LICs) in sub-Saharan Africa (sSA) and Asia where CHB infection is endemic.
The World Health Assembly (WHA), in response to the increasing CHB-related morbidity and mortality, adopted a resolution to improve viral hepatitis prevention, testing and treatment worldwide [4]. Subsequently, the United Nations General Assembly and the World Health Organisation (WHO) respectively incorporated viral hepatitis elimination in the 2030 Agenda for Sustainable Development [5,6] and the WHO global health sector strategy on viral hepatitis 2016-2021 [3,7]. The WHO further recommends hepatitis B surface antigen (HBsAg) testing for all HIV-infected patients, and highly active antiretroviral therapy (HAART) containing at least two drugs effective against hepatitis B virus (HBV) for all HIV-HBV coinfected patients irrespective of disease stage or CD4 count [8], or where not feasible to treat patients with severe liver disease [9].
Whilst both HIV and HBV are endemic in sSA, the prevalence of HIV-HBV coinfection is highest in West Africa [10] where as many as 1 in 10 HIV-infected patients is coinfected with HBV. Uncontrolled HIV-HBV coinfection, due either to lack of or ineffective antiretroviral therapy (ART), increases the risk of death by more than twofold compared to HIV-monoinfected patients [11], irrespective of presence of liver disease [12]. Early and effective Tenofovir Disoproxil Fumarate (TDF) based ART, on the other hand, reduces risk of death among coinfected patients [10,11] underlying the need for implementing universal HBV testing and prompt ART treatment [8].
The Prevention of Liver Fibrosis and Cancer in Africa (PROLIFICA) programme in 2016 found high HIV and HBV prevalence among Gambian communities (2% and 8.5% respectively), with poor knowledge of hepatitis B and limited access to screening and treatment for HBV infection [13]. One of Africa's smallest and poorest nations, The Gambia has one of the highest prevalence of HCC in West Africa [14] with two-thirds of these cases attributable to HBV infection [15]. In order to comply with the recent WHO goal for HBV elimination, we assessed the coverage of hepatitis B testing and treatment, and its impact on clinical outcomes in the largest HIV treatment centre in The Gambia, West Africa.

Statistical analyses
Demographic, clinical and virological characteristics of the study participants were presented according to their HBsAg status, and comparison made between HIV-monoinfected and HIV-HBV coinfected patients using the chi-squared test, Fisher's exact test, or Wilcoxon ranksum test. Factors associated with significant fibrosis (APRI >1.5) were identified using univariable logistic regression. As the number of participants with APRI >1.5 was small, multivariable analysis was not performed.

Ethical approval
The MRC Scientific Coordinating committee and the joint Gambia Government/MRC Ethics committee approved this study (SCC1388). Consenting of participants followed group and individual sensitization, explanation of a study protocol and sufficient time for questions and discussion with spouse/family.

Study population and compliance with guidelines
Between September 2015 and February 2016, 870 HIV-infected patients regularly seen at the Hands on Care HIV clinic were invited to participate in our study and all agreed to take part. The review of patient case notes revealed that only 187 (21.5%, 95% CI 18.8-24.3) had previously been tested for HBsAg, with 23 (12.3%, 95% CI 8.0-17.9) found to be HIV-HBV coinfected. None had a liver assessment based on liver transaminases, ultrasound, measurement of liver fibrosis or liver biopsy. Whilst all 23 patients previously known to be coinfected had been receiving ART, only 6 (26.1%, 95 CI 10.2-48.4) had received a TDF-based regimen.
Of 870 patients accepted for the HBsAg screening, 94 tested positive giving a seroprevalence of 10.8% (95% CI 8.8-13.1). Of these, 15 (16%) were lost to follow up, and 1 died before the liver assessment. The remaining 78 HIV-HBV coinfected patients underwent full liver assessment, along with 40 randomly selected HBsAg negative controls (Fig 1).

Assessment of liver fibrosis
Compared to HIV-monoinfected patients, HIV-HBV coinfected patients had significantly higher median APRI (0.42 vs 0.58, p = 0.002) and FIB-4 (1.72 vs 1.35, p = 0.004), suggesting higher rate of liver fibrosis in coinfected patients. Table 2 shows the factors associated with significant liver fibrosis.
We assessed the joint effect of HBV co-infection and 3TC resistance mutations on significant fibrosis (APRI >1.5) in HIV-infected patients. Compared to mono-infected, the prevalence of significant fibrosis was higher in those coinfected, and especially in those with 3TC resistance mutations. While no HIV-monoinfected patients (0/39) had significant fibrosis, 6/ 66 (9%) of patients coinfected with wild type HBV and 1/8 (13%) coinfected with resistant HBV strain had significant fibrosis (p for trend 0.05).

Discussion
In The Gambia, a West African country highly endemic for HBV infection [13], our study provides two key messages: 1) poor compliance with the WHO hepatitis B testing and treatment guidelines [9] is poor with only 21.5% of HIV patients systematically screened for HBV infection and less than 15% on TDF-based ART, which puts HIV patients at risk of liver complications; 2) testing and linkage to care for HBV infection in HIV patients is accepted and feasible in HIV care programme in The Gambia.
To date, the compliance with the WHO guidelines [8,9] on the management of HBV and HIV coinfection in sSA-a region with high HIV-HBV coinfection prevalence [10], has been poorly documented. We found that in 2016, despite strong international recommendations and increasing global commitment to HBV elimination, HBsAg testing and effective HBV treatment remains poor in the largest Gambian HIV facility predisposing coinfected patients to severe clinical and virological outcomes, increased risk of liver-related morbidity, and diminished benefits of ART. In our study, none of the 870 HIV-infected patients refused to be screened for HBV and almost 85% of HBsAg positive participants accepted to undergo liver assessment. This result is important since it suggests that HBV screening and linkage to care for HBV infection may be widely accepted in HIV-infected patients and feasible in HIV care facilities in sSA. This needs to be confirmed on a larger scale.
Whilst some countries in sSA successfully adopted a TDF-based first-line ART regimen in HIV reference centres [21], indication to treat and choice of ART regimen in Gambian HIV clinics was primarily determined by the type and stage of HIV infection. As a result, 3TC, a cheap and well-tolerated nucleoside analogue has been the backbone of first line ART in The Gambia since its introduction over a decade ago [22]. Although 3TC is an effective antiviral drug, it is associated with an increased risk of HBV resistance mutations following 12 to 24 months of monotherapy in patients with HBV infection [23]. Unsurprisingly, HBV DNA was detectable in almost all our study patients (98.1%) who received prolonged 3TC without TDF.
We found significantly higher prevalence (83.3%) of 3TC associated HBV resistance mutations compared to the 14.2% reported in a similar Gambian HIV-HBV cohort six year earlier who received 3TC without a second anti-HBV drug [24], as well as the 29.3% and 8.7% reported in a Ghanaian [25] and Ivorian [26] coinfected patients respectively. Our findings confirm that widespread prolonged usage of 3TC monotherapy in ART poses diagnostic and management problems to individual patients as well as a potential public health risk. The high proportion of patients with detectable HIV viral load and low CD4 count after many months of ART, as well as the detectable HBV DNA among almost all coinfected patients receiving TDF in our study could be the result of unsatisfactory adherence to therapy, which might contribute to higher levels of HBV viremia and consequently to higher fibrosis and HBV resistance. Another explanation could be the deteriorating HIV services in The Gambia since 2007 when its former president claimed to have found herbal cure for HIV infection. This claim disrupted decades of HIV campaign and negatively impacted vital HIV services and research partnerships resulting in significant reduction in competent personnel and services in HIV facilities and suboptimal HIV care with frequent medication stock outs and several patients having their ARV regimens switched often. These underline the urgent need to improve HIV services and drug provision in Gambian HIV facilities.
HBV co-infection increases liver fibrosis [25,[27][28][29][30][31] as well as both all cause and liver related mortality [11,32], often in the absence of any evidence of liver disease [12]. Although our study lacked statistical power to show definitively a positive relationship between 3TC resistance mutation and increase in liver fibrosis, we show a trend towards such relationship (p = 0.05). Prompt initiation of TDF-based ART has been shown to reduce HBV viral replication and mortality in European [33] and Africa HIV-HBV cohort [11,12], further strengthening the need for compliance with the WHO HBV testing and treatment guidelines in Africa. It is worth noting that the uptake of TDF in first-line ART regimens in sSA is poorly documented. Our study shows that until 2016, only a minority of HIV patients in The Gambia received HBV testing and an even lesser proportion (7/78) of HIV-HBV coinfected patients received TDF. None of the patients had benefitted from a liver assessment even on the basis of liver transaminases. Indeed the local laboratory at this HIV facility, and almost all other HIV facilities in The Gambia, do not have the capacity to assess liver biochemistry. The 2016 consolidated HIV guidelines [8] recommending TDF-based ART for all HIV patients irrespective of degree of liver disease are thus well adapted to resource limited African settings, and African HIV facilities need to be supported to implement these guidelines.
Our study has a few limitations: first, we analysed a limited number of HIV-HBV coinfected patients; second, we assessed liver fibrosis using APRI score as recommended by the WHO for resource-limited settings without access to Fibroscan, which is the case for most HIV facilities in Africa where Fibroscan is rarely available outside research settings. The reliability and usefulness of platelet-based biomarkers like APRI for assessing liver fibrosis in HIV patients in Africa has been questioned [34] mainly because thrombocytopenia is commonly associated with advanced HIV infection. Third, our study did not assess for hepatitis C virus (HCV) and hepatitis D virus (HDV) coinfection. However, previous studies in The Gambia confirmed very low prevalence rates of both HCV and HDV infections [13]. Finally, our study suggests that adherence to ART was not satisfactory but we did not directly assess adherence to ART.

Conclusion
Compliance with international HBV testing and treatment guidelines for HIV-infected patients is poor in this Gambian HIV programme, resulting in a low coverage of HBV testing, prolonged treatment with 3TC without TDF for coinfected patients, and poor HBV viral suppression with increased risk of HBV resistance and liver complication. However, interventions to improve HBV screening and linkage to care for HBV infection in HIV facilities are feasible.
Supporting information S1 File. Data underlying the findings described in the manuscript. (XLSX)