D28 HIV and hepatitis co-infections

Introduction

With the advance of highly active antiretroviral therapy (HAART), HIV infected people are living longer. While they are less likely to be dying from opportunistic infections, one of the main causes of death today is complication from chronic liver disease, especially chronic hepatitis B virus (HBV) and/or hepatitis C virus (HCV) diseases. Since HIV, HBV and HCV share similar transmission routes, it is not uncommon to see HIV patients co-infected with either HBV or HCV or even both viruses. It is estimated that among the 36.7 million HIV infections worldwide, 2.7 million are co-infected with HBV and 2.3 million are co-infected with HCV. However, there are significant geographical variations. In Hong Kong, the HBV infection rate in the HIV population can be up to 16% and the HCV co-infection rate may reach 25% in some sub-populations. Co-infection within the community pf people who inject drugs (PWID) is almost universal.

HBV and HIV

Interaction between HBV and HIV

It is estimated that about 257 million people are chronically infected with HBV globally and about 0.9 million of them die annually. The prevalence is especially high in Asia and sub-Sahara Africa, where the infection is commonly acquired during perinatal period or early adulthood. The cumulative lifetime incidence of cirrhosis and hepatocellular carcinoma (HCC) in mono-infected HBV patients was reported as 41.5% and 21.7% respectively.[1] Risk factors for progression include increasing age, seropositive hepatitis B e antigen status (HBeAg), male gender, increasing HBV DNA and alanine transaminase (ALT) levels, and genotype C.

HIV has a strong tropism for liver stellate cells and hepatocytes through its chemokine co-receptors and is able to exert a direct cytopathic effect on the liver cells. Studies have shown that HIV-HBV co-infected patients have more morbidities and higher mortality than those caused by either infection alone. Compared to mono-infection, they have higher HBV DNA level, 5 times faster progression to chronic hepatitis B after acute HBV infection (persistent positive HBsAg 6 months after acute HBV infection), higher risk of cirrhosis and HCC.[2] Moreover, they have lower HBeAg seroconversion rates but a higher risk of HBeAg and hepatitis B surface antigen (HBsAg) seroreversion (loss of anti-HBe and anti-HBs with reappearance of HBeAg and HBsAg). Liver related mortality is about 18 times higher in co-infected patients compared with those with HBV mono-infection. The mortality rate is even doubled after the introduction of HAART, which may be related to the unmasking effect of liver diseases following a reduction of AIDS associated deaths.[3] Cohort studies have suggested a slower virologic response to HAART in the presence of HBV infection, while others reported no significant impacts on HIV disease progression.

Challenges in the management of HBV-HIV co-infection

The main challenges in the management of HIV-HBV co-infection are the presence of occult HBV infection, immune reconstitution and drugs related issues.

Occult HBV infection is defined as HBV viraemia in the absence of detectable HBsAg. In some patients, anti-HBc and/or anti-HBs could be present. The impact of occult hepatitis B is still controversial but there could be associated progression to cirrhosis and HCC. The prevalence of occult hepatitis B could be as high as 10% in the HIV population, especially when the CD4 cell count is <200 cells/μL. Testing for anti-HBc and possibly HBV DNA should also be considered when occult HBV infection is suspected.

Immune reconstitution inflammatory syndrome (IRIS) is associated with immune recovery after the initiation of HAART (refer to Chapter C17). Since HIV infection can lower HBV specific T-cell response, the control of HIV replication can enhance T-cell function and thus HBV control. As a result, reduction of HBV DNA and enhancement of HBV seroconversion may be seen even before the initiation of HBV specific therapy in co-infected patients. However, hepatitis flares can occur after immunologic recovery in patients with limited liver reserve.

On the treatment of chronic HBV infection, there are currently 7 oral antivirals (lamivudine (3TC), emtricitabine (FTC), telbivudine (LdT), adefovir (ADV), entecavir (ETV), tenofovir Disoproxil fumarate (TDF) & tenofovir alafenamide (TAF)) and 2 injection therapies (standard interferon alpha (IFN) and pegylated interferon alpha (pIFN)). Some of these have anti-HIV activities: 3TC, FTC, TDF, TAF and ETV. Prescription of the latter anti-HBV medications to co-infected individuals should be avoided since resistance to HIV would develop shortly afterwards if HAART is not given simultaneously.

HCV and HIV

Interactions between HCV and HIV

Worldwide about 2% of the world population is estimated to be chronically infected with HCV infection. Sharing needles in PWID is the most common route of transmission in most countries, In Hong Kong over 80% of PWID are HCV antibody positive. HCV can be divided into 6 major genotypes with over 50 subtypes. Genotype 1 is commonest in the United States and Europe, while genotype 3 is more common in India, Australia and the Far East. Genotype 4 is prevalent in Africa and Middle East, and genotype 5 in South Africa. In Hong Kong, the most common genotypes are 1b and 6a,[4] with a higher prevalence of genotype 6a in PWID. Sexually acquired HCV in HIV patients appear to cluster with other genotypes.

Following acute HCV infection, about 20-40% of mono-infected patients and only 5-15% in HIV/HCV-coinfected patients can clear the virus. The HCV RNA level in HIV co-infected patients is also consistently higher than that of non-HIV individuals. This may somehow translate into enhanced HCV transmission.

HIV infection has been shown to accelerate HCV-associated liver fibrosis, increase risk of hepatic decompensation, HCC and death. The risk factors for hepatic fibrosis include age over 25 years at acquisition of HCV, low CD4 cell counts, use of alcohol >50g daily, baseline necroinflammatory activities and detectable HIV RNA in serum.[5] However, HAART has been shown to be associated with lower risk of end stage liver disease, all-cause mortality and HCC. Similar to HBV-HIV co-infection, the effect of HCV infection on the progression of HIV infection is also conflicting. A large cohort study has demonstrated that HCV-HIV co-infected individuals had a 70% higher chance of progression to AIDS or death, and was associated with a lower CD4 recovery after HAART.[6]

Challenges in the management of HCV-HIV co-infection [Further reading C]

While both HIV and HCV are viral infections which should benefit from antiviral therapy, there are the challenges of balancing the varied treatment strategy, concerns for interactions with HAART, and the consideration of acute infection and reinfections.

Unlike the lifelong regimens for HAART, HCV therapy is time-limited aiming at viral clearance. HCV treatment of co-infected patients is modelled on that for mono-infection. Some of the main considerations of HCV therapy are:

1. With the use of pIFN-based therapy, it is necessary to differentiate between ‘easier-to-treat’ and ‘harder-to-treat’ genotype. In Hong Kong, the former refers to genotype 2, 3, 6, while the latter 1, 4, 5. However, with the availability of new Direct Acting Antivirals (DAA), all genotypes can potentially be treated effectively.
2. The application of virological response-guided therapy is used in the conventional pIFN-based therapy, with the monitoring of the following response markers:
   • SVR (sustained virologic response) – the absence of HCV RNA 12 (SVR12) or 24 (SVR24) weeks after the end of treatment. SVR12 is the main goal of therapy for DAA therapy, while SVR24 is the goal of treatment for pIFN-based therapy.
   • RVR (rapid virologic response) – undetectable HCV RNA 4 weeks after initiation of treatment
   • EVR (early virologic response) – undetectable HCV RNA (complete EVR) or more than 2 log drop in HCV RNA (partial EVR) at week 12. This marker is used for pIFN-based therapy only.
   • ETR (End of treatment response) – undetectable HCV RNA at the end of therapy.

As regards the interaction with HIV therapy, chronic HCV infection may increase the hepatotoxicity of antiretrovirals although HAART itself can slow down hepatic fibrosis progression. Overall, the relative risk for liver enzyme elevation after starting HAART is 2.78 in co-infected patients compared to those with HIV infection alone. However, the clinical benefit of HAART outweighs the risk of liver damage. In practice, HAART regimens with profound hepatotoxicity are best avoided, and modification of regimens is usually not necessary.

Finally, while chronic HCV infection has been the focus of HIV management, the occurrence of acute HCV infection should not be ignored. Since 2013 locally, there has been an increase in the number of reported acute HCV infection, predominantly genotype 3, as a result of sexual transmission in men having sex with men (MSM). On the other hand, re-infection could have been under-estimated, which should be considered following effective HCV therapy.

Management of viral hepatitis/HIV co-infections

All HIV infected patients should be screened for HBV and HCV infections. ALT levels, HBsAg, anti-HBs, anti-HBc and anti-HCV Ab should be tested annually. HBV DNA should be determined quantitatively if HBsAg is positive. Similarly, HCV RNA and genotype should be tested if anti-HCV is tested positive. HBV DNA and/or HCV RNA levels may be considered in high-risk individuals with unexplained elevated ALT level as there may be occult infections. HBeAg/anti-HBeAb and HBV DNA should also be checked in HIV-HBV co-infected patients at regular intervals to monitor for seroconversion and virologic response to anti-HBV treatment, particularly when immunosupression is reversed with the use of HAART.

In view of the increased risk of developing HCC and cirrhosis, co-infected patients should have regular screening with alpha fetal protein and abdominal ultrasound scan at 6 to 12 months intervals. Computer tomography scan or magnetic resonance imaging of the liver may also be required to examine for space occupying lesions in case of doubt. The degree of the hepatic necroinflammation and fibrosis can be better assessed by performing a liver biopsy. Histology can also provide information on comorbid conditions like non-alcoholic steatohepatitis or even granulomatous infections. However, because of the potential risk of the procedure and, with the increasing use of non-invasive fibrotic assessment tool like transient elastography (Fibroscan^®), liver biopsy is becoming less favoured nowadays, although it remains the gold standard for assessment.

Since hepatitis-HIV co-infected patients are prone to more severe liver damage, they should be advised to refrain from alcoholic intake. Hepatotoxic medications should be avoided as far as possible. In view of the common route of transmission among the 3 viruses, patients should be advised to refrain from sharing intravenous needles and to practise safer sex with condom use.

Vaccination against hepatitis A virus (HAV) should be considered if one’s anti-hepatitis A IgG antibody is screened negative. For HCV-HIV co-infected individuals, HBV vaccination should also be considered. A standard 3 dose (at 0, 1 and 6 months) or the 4 dose schedules (at 0, 1, 2 and 12 months) have been shown to be effective in HIV patients. Post-vaccination antibody level can be assessed 1-2 months afterwards. Patients with anti-HBs titre lower than 10 IU/L can be given a full course or a booster dose. Patients failing to seroconvert despite repeated vaccinations are considered at risk, and HBsAg and anti-HBc should be monitored annually. In general, patients with HIV are known to have suboptimal response to HBV vaccination. Poor responders are typically associated with a higher baseline HIV viral load and lower CD4 cell count. The response in patient with CD4 ≥350/μL has been shown to be more than 3 times higher than those with counts <350/μL.[7] Therefore HBV vaccination may be postponed until there is immunologic recovery after the initiation of HAART. Nowadays, there is no vaccination for HCV infection.

Management of HIV-HBV co-infections [HK Guidelines 28A]

In the management of HIV infection, cautious use of antiretrovirals in co-infected patients is advised since some of the compounds are hepatotoxic. Didanosine (ddI) has been shown to be associated with portal hypertension, steatohepatitis and liver fibrosis. Nevirapine (NVP) was also linked to liver fibrosis and elevated liver enzyme besides hypersensitivity. Other antiretrovirals with hepatotoxic effects are tipranavir (TPV), darunavir (DRV), stavudine (d4T), ritonavir boosted lopinavir (LPVr), fosamprenavir (FPV), while atazanavir (ATV) and efavirenz (EFV) should also need to be used with caution.

As regards to HBV, the goals of antiviral treatment are to suppress viral replication (HBV DNA undetectable), improve immune control, normalise liver enzymes and histology, so as to prevent progression to end stage liver disease and HCC, and finally to eradicate the virus. The latter objective is difficult to achieve and therefore a long term and even lifelong therapy is usually needed. The dosages and characteristics of the anti-HBV medications are listed in Box 28.1.

It is agreed that all patients with HIV/HBV coinfections should have their HIV treated with antiretroviral regimens including TDF/TAF + 3TC or FTC. Use of TAF instead of TDF is recommended in patients with established renal disease with a creatinine clearance >30mL/min or osteoporosis.

The preferred combinations TDF/TAF + FTC or TDF/TAF + 3TC should be used as the NRTI backbone of a HAART regimen regardless of the CD4 cell count. TDF/TAF are highly potent against HIV and HBV. HBeAg seroconversion and HBV DNA suppression can be achieved when TDF/TAF is combined with 3TC or FTC. It can also suppress 3TC resistant HBV infection. TDF/TAF resistance to HBV is rare. However, if TDF/TAF cannot be used for some reasons, ETV can be used on top of a fully suppressive HAART regimen. For 3TC or FTC resistant HBV infection, double dose ETV (1mg daily) should be used instead. Though 3TC and FTC are potent drugs for HBV infection, they should be used in conjunction with TDF/TAF because of the high likelihood of resistance to HBV (M204V mutation in 3TC at 25% per year).[8] Other antivirals are less commonly used in HIV/HBV coinfection. Co-infected patients generally respond poorly to pIFN and its use is associated with higher degree of toxicity. ADV has minimal HIV activity and does not create problem when used alone for co-infected individuals. Its relatively inferior potency and higher risk of renal toxicity limits its widespread use. LdT has higher efficacy against HBV infection than 3TC and ADV. There is however moderate risk of resistance on prolonged use, and cross resistance to 3TC/FTC.

Management of HIV-HCV co-infections

The goal of HCV treatment is to achieve SVR. SVR has been shown to be associated with decreased mortality, reduction of liver related complications including HCC and need for liver transplantation. HCV treatment should be initiated before HAART to minimise potential drug-drug interaction (DDI). However, the decision to treat must rest on the consideration of both the HIV and liver states:

1. HAART should be started for all co-infected patients regardless of the CD4 cell count.
2. HCV treatment should be considered for all co-infected patients regardless of their liver status.
3. All co-infected patients should be screened for hepatitis B. Patients who are not immune to hepatitis B should be given a complete course of hepatitis B vaccine.
4. HBV reactivation may occur in HBV infected patients given DAA therapy. Therefore, HCV/HIV co-infected patients with active hepatitis B should be treated with HAART that includes 2 agents with anti-HBV activity prior to HCV therapy.

Pegylated Interferon (pIFN)-based regimen:

The conventional treatment for HIV/HCV co-infection comprises the use of pIFN and Ribavirin (RBV), same as for HCV mono-infection, with modifications.[Algorithm 28A] One important consideration is that the SVRs for co-infection are generally lower than HCV infection alone for all genotypes: genotype 1: 17-35%; genotype 2/3: 44-72%.[9] The predictive value of RVR and EVR is likewise different for co-infected patients.

Patients with genotype 1, 4 or 5 should be treated for 48 weeks if there is RVR, or 72 weeks if there is partial EVR and negative HCV RNA at 24 weeks of treatment. Genotype 2, 3 or 6 can be treated for 24 weeks if RVR is achieved, otherwise 48 weeks are needed. Two of the most significant predictors for SVR are genotypes and baseline HCV RNA. Higher SVR can be seen in patients with lower baseline HCV RNA (≤400,000 to 800,000 IU/mL), non-African Americans race, younger age (<40 years old) and lower body weight. Besides, the IL28B gene polymorphism, is a strong predictor for SVR to HCV treatment. The CC genotype is significantly associated with higher SVR. Since East Asians have higher rate of CC genotype than European, Americans and Hispanics, they tend to have a better SVR.

An SVR of more than 70% can be achieved with a complete EVR and the treatment duration can be kept at 48 weeks. In those with partial EVR and negative viral load at 24 weeks, the treatment can be prolonged to 72 weeks. Treatment should be stopped if EVR is not achieved. Complete blood counts and liver function tests should be monitored at 2-4 weeks interval. CD4 and thyroid function tests should be monitored at 12 weeks interval.

pIFN-based regimens have a lot of adverse events. pIFN may give rise to flu-like symptoms, cytopaenia, autoimmune activation, thyroid dysfunction and mood change. RBV can cause headache, nausea/vomiting, anaemia, teratogenicity, neutropaenia, insomnia and depression. Tenofovir based NRTI for HIV therapy is preferred in co-infected patients given pIFN-based therapy in view of the aggravation of anaemia by zidovudine and possibly lowering of SVR with abacavir.

Because of the availability of DAAs, adverse events and lower efficacy of pIFN-based regimens, they are no longer recommended in many international guidelines.

Direct Acting Antivirals (DAA) for HCV/HIV coinfection [Box 28.2][Box 28.3]

Development of DAAs has revolutionised the management of chronic hepatitis C by having pIFN-free treatment in place, shorter duration of therapy, lower rates of side effects and higher efficacy. The SVR12 response to DAA-based therapy is almost 100%. It is therefore anticipated that eradication of HCV is becoming achievable. In contrast to pIFN/RBV therapy, the use of DAAs in HIV/HCV co-infection gives comparable SVR rates as that in HCV mono-infection. Therefore, indications for treatment are the same for mono-infected and HIV/HCV co-infected patients. Nevertheless, regimen consideration for the co-infected will depend on genotypes, prior treatment experience, fibrosis status and drug interactions with ARTs.[Box 28.3] DAAs can be classified into the following classes:

1. NS3/4A Protease inhibitors (PI): These compounds inhibit the NS3/4A serine protease, an enzyme involved in post-translational processing and replication of HCV. In general, NS3/4A PI inhibitors have high potency but moderate genetic barrier to resistance and multiple DDI. 1st generation PI including Boceprevir (BOC) and Telaprevir (TVR) were withdrawn from the market because of their side effects and the availability of the more potent newer DAAs. 2nd generation PI have significantly better SVR compared to 1st generation PI with fewer side effects, less DDI, and the need for a shorter duration of treatment. Examples including simeprevir (SMV), asunaprevir (ASV), paritaprevir (PRV) & grazoprevir (GRZ) can be used in genotypes 1 & 4, while glecaprevir (GLE) & voxilaprevir (VOX) are pan-genotypic.
2. NS5A inhibitors: They inhibit viral replication and assembly of HCV. In general NS5A inhibitors are highly potent but have low genetic barrier to resistance. Therefore they should not be used alone. Examples include: daclatasvir (DCV), ledipasvir (LDV), elbasvir (EBR), ombitasvir (OBV) and velpatasvir (VEL) & pibrentasvir (PIB).
3. NS5B inhibitors: They inhibit RNA synthesis at different stages/sites resulting in inability of viral RNA replication. Sofosbuvir (SOF) is the first member of the nucleoside NS5B inhibitors being developed. It has intermediate potency towards HCV but a high genetic barrier to resistance. It is pan-genotypic and is highly potent when combined with other DAAs. Dasabuvir (DSV) is a non-nucleoside NS5B inhibitor. However, in contrast to SOF, it has low genetic barrier to resistance and limited efficacy for genotypes other than 1.

Genotype based regimens [Box 28.2]

Effective DAA regimens contain combination of agents from different classes. Most recommended regimens last for 12 weeks. Shortened duration (8 weeks) with use of SOF/LDV is possible in treatment-naive genotype 1 patient with no cirrhosis and low baseline viral load (<6 million IU/ml). Other regimens for genotype 1 include SOF/DCV, EBR/GRZ, 3D & ASV/DCV. The first 3 regimens last for 12 weeks in treatment-naive non-cirrhotic patients. ASV/DCV is given for 24 weeks. Resistance associated substitutions (RAS) for DCV are frequently observed at position 31 and 93 in genotype 1. SVR for ASV/DCV treatment is significantly lowered with the presence of RAS. Therefore prior assessment of RAS is mandatory for this regimen. However, evidence on ASV/DCV treatment for HCV/HIV co-infected patient is limited. Genotype 1a is considered to be more difficult-to-treat than genotype 1b. The addition of weight-based RBV to either EBR/GRZ or 3D (PRV/ritonavir + OBV + DSV) can improve the SVR. Extension of duration of treatment may be required in EBR/GRZ + RBV regime if the baseline HCV RNA is greater than 800,000 IU/ml or with the presence of RAS.

Genotype 2 and 3 can be treated with the pan-genotypic regimens including GLE/PIB, SOF/VEL and SOF/DCV. Extension of duration of SOF/DCV treatment to 16 weeks to 24 weeks for genotype 2 is required in patients with Child A cirrhosis. For genotype 3, extension of SOF/DCV treatment duration to 24 weeks with RBV is needed in view of its difficult-to-treat nature.

Genotype 4 is rarely seen in Asia including Hong Kong. Regimens including SOF/DCV, SOF/LDV, SOF/VEL, EBR/GRZ, 2D + RBV can be used.

Genotypes 5 & 6 are considered relatively easy-to-treat genotypes. Regimens including SOF/LDV, SOF/VEL & GLE/PIB can be used.

Pan-genotypic regimens

There are 4 pan-genotypic regimens: GLE/PIB, SOF/VEL, SOF/DCV & SOF/VEL/Voxilaprevir(VOX) (Vosevi^®). GLE/PIB can be given for 8 weeks in either treatment-naive or -experienced non-cirrhotic patients, except for treatment-experienced genotype 3 patients of which 12 weeks is required. It has to be given for 12 weeks for cirrhotic patients, however. Twelve weeks of SOF/VEL is also recommended for all patients with any genotypes and liver status. The only exception is genotype 3 patients with cirrhosis. SOF/VEL/VOX can fill the gap in this regard. In general, SOF/VEL/VOX can be given for 8 weeks for treatment naive non-cirrhotic patients and 12 weeks in patients with Child A cirrhosis. It is also recommended as salvage therapy for those who has failed the other DAA regimens and has to be given for 12 weeks.

Drug-Drug Interaction (DDI) between DAA and HAART [Box 28.3]

Many DAAs, especially PIs, are metabolised via cytochrome P450 and thus can potentially interact with antiretroviral medications. It is highly recommended that, prior to the DAA treatment, assessment using guidelines or web-based interaction programs or applications can help to minimise DDI (University of Liverpool: http://www.hiv-druginteractions.org and www.hep-druginteractions.org). Occasionally, HAART regimen may need to be switched to those with minimal DDI, e.g. dolutegravir (DTG) or raltegravir (RAL), before DAA initiation.

SOF has minimal DDI with HAART. LDV has fewer DDI than DCV. DCV can increase Atazanavir (ATV) and Elvitegravir (EVG) levels. The dose of DCV should be reduced to 30mg daily. However, DCV level can be decreased by EFV and thus its dose should be increased to 90mg daily. LDV/SOF can increase TDF level, and thus the risk of renal toxicity. This combination should be avoided in patients with eGFR <60ml/min. In general, 3D, GLE/PIB & EBR/GRZ have multiple DDIs with HIV medications including PI and non-nucleoside reverse transcriptase inhibitors (NNRTI). Therefore, their combination should be avoided. Patients on ritonavir sparing HAART should not be given 3D regimen as, the low dose ritonavir within the 3D preparation may select for HIV PI resistance. Moreover, for those taking ritonavir boosted PI HIV regimen, the addition of 3D may increase the ritonavir level and thus the substantial gastrointestinal side effects. For NNRTI, only rilpivirine (RPV) has no DDI with most DAAs. HIV PIs have no interaction with LDV. For entry or integrase inhibitors, DTG and RAL are free of interactions with all current DAA regimens.

Management of acute HCV infections in HIV patients [Algorithm 28B]

Acute hepatitis C infection is defined that occuring in as the first 6 months after exposure. About 20-40% of the HCV mono-infected can clear the virus spontaneously. However, this will drop to around 5-15% in HIV/HCV co-infected patients. The diagnosis of acute HCV infection entails a positive HCV antibody test after a prior negative test (seroconversion) together with a positive HCV RNA. It was shown that a decrease in HCV RNA by less than 2 logs at 4 weeks compared with initial HCV RNA, and persistent detectable HCV RNA at 12 weeks after the diagnosis of acute hepatitis C, was less likely to result in spontaneous clearance. Treatment can be offered in these 2 groups of patients. However, in patients with high risk of transmission, treatment should be considered at diagnosis. In the past, satisfactory cure rate was achieved with IFN-based therapy for acute hepatitis C infection. With the availability of DAAs, IFN-based therapy is no longer recommended in various guidelines. DAA regimens for acute hepatitis C is the same as that of chronic hepatitis C.

References

Further reading

Hong Kong Guidelines