Population

1. Risk-based screening in different high-risk populations: Injecting drug users (IDUs), men who have sex with men (MSM), immigrants, recipients of blood transfusion and blood products, sex workers, and health-care workers (HCW), HIV-infected persons
2. General population (excluding blood donors) or selected subpopulations of general population (women during pregnancy, those with raised alanine aminotransferase [ALT], Infants, schoolchildren and adolescents)
3. Other approaches: Birth cohort screening (based on different age cut-offs, born between 1945– or 1965 or 1970).
4. One off screening vs repeat screening every five years

Intervention

Testing strategies for HBV in different populations (risk based and general population and birth cohort); and at different prevalence thresholds

Comparator

No testing or current practice or comparison of different testing strategies

Outcomes

Benefits, harms and costs, and cost–effectiveness with different screening strategies for different target populations

Individual patient outcomes: No. of cases detected, overall mortality, liver-related mortality, cirrhosis, end-stage liver disease, rate of hospitalizations, serious adverse events, quality of life

Prevention: New infections (mother to child, horizontal [IDUs needle sharing and sexual; and sexual, esp MSM])

Cost–effectiveness: Cost and incremental cost per case diagnosed; cost and incremental cost per case screened and treated; cost and incremental cost per life saved; cost and incremental cost per infections averted; quality-adjusted life years (QALYs) gained

Background

Epidemiology

Chronic hepatitis B (CHB) – defined as persistence of hepatitis B surface antigen (HBsAg) for six months or more – is a major public health problem. Worldwide, there are an estimated 250 million chronically infected persons, particularly in low- and middle-income countries (LMICs). Universal hepatitis B immunization programmes that target infants, with the first dose at birth, have been highly effective in reducing the incidence and prevalence of hepatitis B in many endemic countries. However, these programmes will not have an impact on HBV-related deaths until several decades after their introduction.

The major complications of CHB are cirrhosis and hepatocellular carcinoma (HCC). Between 20% and 30% of those who become chronically infected will develop these complications, and an estimated 650 000 people will die annually due to CHB.

The risk of developing chronic HBV infection decreases with age at infection, from about 90% when infected perinatally up to 6 months of age to 20–60% between the ages of 6 months and 5 years. Of those who acquire HBV as children 25% will develop primary liver cancer or cirrhosis as adults.

Routes of transmission worldwide: In sub-Saharan Africa and east Asia, transmission predominantly occurs in infants and children by th eperinatal and horizontal routes (i.e. resulting from close contact that is not parenteral, perinatal, or sexual in nature) whereas in more industrialized countries, rates of new infection and acute disease are highest among young adults and transmission predominantly occurs via Injecting drug use and other high-risk behaviours. Worldwide, the majority of infections are acquired at birth or in early childhood.

Low rates of diagnosis: The majority of people are unaware of their HBV infection, and therefore often present with advanced disease. At present, there is a massive burden of undiagnosed and untreated hepatitis B and C, with 40–85% of infected persons undiagnosed, but varies greatly by setting. By contrast, the estimated awareness of status among people living with HIV (PLHIV) is within 40%–60% range for two thirds of countries, but varies significantly (CHAI, UNAIDS Info).

–

Based on still limited studies, overall <15% of the estimated 180 million who are chronically infected with HCV are aware of their diagnosis, based on data from higher-income setting – United States, Europe and China.

–

And from a survey in the US, a similar proportion of those with chronic HBV infection are aware of their diagnosis.

–

The proportion in low-income settings is even higher, with only a tiny fraction diagnosed and aware.

Reasons for low uptake of testing are multifactorial, and include lack of awareness at all levels, lack of clear guidelines, competing health-care priorities, limited health-care budgets and political will. This leads to many people remaining undiagnosed until the later stages of the disease, when prognosis is poor.

In addition to the very low access to and uptake of testing, there is also further attrition in the care cascade with very poor linkage to care and therefore treatment, among those who test positive.

Hepatitis B and C testing and diagnosis are at the core of entry to both the prevention and treatment cascade.

–

Testing is required to identify those with are positive, linking them with care, counselling them on measures to reduce transmission to others then assessing who needs treatment, initiating treatment, achieving treatment response (sustained virological response [SVR] for hepatitis C) or long-term viral suppressi on for HBV and retaining in care for HBV.

–

Hepatitis testing is also needed to identify those who are negative, to provide hepatitis B vaccination, and the opportunity to implement individual or facility-level prevention measures, counsel to reduce risk behaviours, or institute facility-level prevention measures on measures to acquisition.

There are three key approaches to screening

1. Population- or community-based screening (including antenatal). This means that all members of the population have access to the screening programme under consideration. It may also include home-based testing (house to house); campaigns (e.g. HTC plus – malaria, safe water, noncommunicable diseases e.g. diabetes and hypertension); outreach (mobile) in general and key populations; workplaces and schools; and health-care facility0based screening.
2. Health-care facilities. Testing could also be offered in special dedicated clinics, e.g. HIV, STI clinics. Screening at health-care facilities may include primary care settings, inpatient and outpatient settings, and may involve screening on the basis of clinical presentation or focus on only those with abnormal liver function tests, abnormal ultrasound scan, family history of liver disease or other clinical suspicion of liver function test.
3. Targeted risk factor-based screening. This refers to screening of specific groups including key populations, who are generally at higher risk of being infected than the general population. This includes people who inject drugs (PWID), people in prisons and other closed settings, migrant populations, some indigenous populations, MSM and sex workers, but may also include health-care workers. People attending services providing care and treatment for viral hepatitis or HIV can be encouraged to bring their partners to be tested.
4. Birth cohort screening for HBV and HCV.

Existing guidelines: what are countries doing?

1. Most countries have based their list of high-risk groups as defined by the Centers for Disease Control and Prevention (CDC), and are largely based on known modes of transmission. Generally they include recommendations for three main screening approaches: (Apata, MMWR Morb Mortal Wkly Rep. 2014;63:613–19; Weinbaum, Hepatology. 2009;49:S35–S44; Han, Vaccine. 2013;31 Suppl 9:J36–J42)

   • Population-based screening that includes antenatal clinic screening
   • Screen those with high-risk behaviours, exposures and other conditions

     –

     Family members and household contacts of hepatitis B patients

     –

     MSM

     –

     PWID

     –

     HIV-positive patients

     –

     Patients on immunosuppression or chemotherapy

     –

     Persons with liver disease of unclear etiology

     –

     Health-care workers.

   • Birth cohort for HCV screening in US and Japan.
2. At present, there is no universally accepted recommended screening programme. There is widespread testing of blood donors (but not necessarily universal), and widespread antenatal screening and infant vaccination in Asia. In addition, there is a risk factor-based testing in highrisk groups in Asia (PWID, liver disease, renal dialysis) and use of a birth cohort approach in the US and Japan.

Survey of guidelines (Surjo De)

Evidence: systematic reviews of prevalence of HbsAg

1. General population: systematic review (Ott, Lancet 2015)

161 countries included.

High endemicity (>5%): Most countries in Africa were of higher–intermediate endemicity (HBsAg prevalence 5–7·99%), or highly endemic for HBV (HBsAg prevalence ≥8%. The Western Pacific Region was also a high–intermediate endemicity region (5–7.99%), especially in the Pacific Island States such as the Solomon Islands.

Intermediate endemicity (2–5%): The Eastern Mediterranean Region was of lower–intermediate endemicity (2·00–4·99%), but Djibouti, Somalia and Sudan showed a higher prevalence of HBsAg than other countries in the region such as Iran.

Low endemicity: Countries in the Americas, such as Mexico, Guatemala, and the USA had mostly low endemicity levels (HBsAg prevalence <2%), ranging from 0.01% (95% CI 0.01–0.01) in the UK to 10.32% (8.56–12.38) in Kyrgyzstan. Overall, the South-East Asia Region had low endemicity levels but on country level, an HBsAg prevalence below 2% was only noted in India, Indonesia and Nepal.

Summary of prevalence across risk groups

2. PWID: Systematic review ((Nelson et al. 2011))

PWID are a key population who are at particularly high risk of HBV, HCV and HIV infection. In many high-income countries and some developing countries, ongoing HCV transmission is driven mainly by PWID populations. A review of global prevalence data from 77 countries estimated that exposure to HCV (anti-HCV positive) among PWID is estimated to be between 60% and 80% in 25 countries, and over 80% in 12 countries. Similarly, of 59 countries where data were available, prevalence of HBsAg among PWID ranged from 5% to 10% in 21 countries and over 10% in 10 countries.

PWID data are global by 20 Global Burden of Disease regions

Southeast Asia and East Asia: 2.9–19.5%

South Asia: 5.8–17.3%

Central Asia: 7.9%

Eastern Europe: 0.5–21.3%

Central Africa: 3.8–9.0%

Andean Latin America: 2.3–8.6%

3. MSMs and sex workers: systematic review ((Hope et al. 2014)))

MSM can acquire HBV and HCV sexually. In many populations, there are higher rates of HBV infection among MSM, requiring targeted HBV screening and vaccination. MSM who are HIV positive are at significantly higher risk of acquiring HCV infection than HIV-negative MSM.

Sex workers are a key population who are at high risk of acquiring HBV and HCV infection. Multiple factors may contribute to this vulnerability, including unsafe working conditions, barriers to negotiating consistent condom use, and difficulties accessing health-care services.

MSM (12 countries), sex workers (5 countries)

View in own window

Country	MSM	Sex workers
Albania	18%
Azerbaijan	4%	3.3%
Croatia	0.9%	1.4% (Bosnia)
Georgia	10%	11.1%
Serbia (incl. Kosova)	8.7%	18.3%
Turkey	3.6%	2.4%
Ukraine	9.8%	9.1%

4. Migrants and refugees: systematic review (Rossi et al. 2012)

sub-Saharan Africa Region: 10.3%

East Asia and Pacific: 11.3%

Central and Eastern Europe and Central Asia Region: 5.8%

5. HIV-infected persons: systematic review (Easterbrook et al. 2015)

HIV/HBV (483 estimates from 75/193 (39%) countries

View in own window

	Gen pop	PWID	MSM	Hetero	Pregnant
	Mid-point co-infection prevalence (Interquartile range) Number of studies
East Africa	8% (6–11) 10		9% 1	6.5% (5–10) 10	4% (2–5) 2
West, Central Africa	11% (6–15) 11		22% 1	12% (8–20.5) 32	9% (0–13) 3
South Africa			6.5% 1	7% (5–20) 7	5% (3–6) 2
Latin America	1% (0.6–2) 3	27% 1	9% (6–11) 5	3% (2–7) 4	0.5% (0.5–1.8)3
North America		7% 1	5% (5–6) 2	17% 1
South East Asia	2% (1–2)2	18% (10–20) 10	15% (10–19)6	9% (0–15) 10
Eastern Europe and CAR
Europe		4% (3–7) 3	5% (4–6) 9	7% (2–11) 2
East Med	10% 1	8% (4–44) 6
East Asia		9.5% (2.5–37)4	12% (10–13.5)4	5% (4–6%) 4
Western Pacific			4% (3–5) 6

HBsAg prevalence based on a total of 170 estimates in HIV-infected persons, based on population type (general population, PWID, MSM, heterosexual, and pregnant women) and by eleven geographical regions.

1. First, reflecting the epidemiology of HBV in Africa whereby the majority of HBV infections are acquired perinatally or in childhood, the prevalence among key populations of HIV-infected PWID and MSM is not substantially higher than the background rate in the general population or among heterosexuals, especially in Africa,
2. Only in the South-East Asia Region is there a higher prevalence in among PWID and MSM.

6. Prisoners

The prevalence of HBV in prisons is often significantly higher than in the general population. Globally, the prevalence of HIV, STIs, hepatitis B and C and tuberculosis in prison populations is estimated to be two to ten times higher than in the general population, and in some settings, 50 times higher. People in prisons and closed settings may be at particular risk for HBV, HCV and HIV infection for a number of reasons. Most commonly, this is due to sharing of needles and syringes and other injecting equipment; often because prevention hardware such as clean needles and syringes are not accessible to prisoners.

7. Indigenous populations

In some settings, indigenous populations are also disproportionately affected by viral hepatitis infection, along with a number of other health problems. Contributing factors to these disparities may include higher rates of injecting risk behaviours among indigenous people who inject drugs and higher rates of incarceration.

Guiding principles for hepatitis B and C testing

1. Promotion of health equity and human rights in national hepatitis B and C testing so that: expanded testing and access is fair and equitable; priority for testing is on diagnosing the undiagnosed; identifying those in greatest need of treatment and those with ongoing risk of infection; and that testing is voluntary and care is provided in a supportive environment free of stigma and discrimination.
   This is critical as many of the affected population are those who are systematically excluded from access to testing, treatment and care, such as sex workers, injection drug users, men who have sex with men, and prisoners.
2. All persons who test positive for hepatitis B and C (in addition to HIV) should have access to and be linked to hepatitis care and treatment services.
3. Testing of key populations should be undertaken where possible in conjunction with other risk or harm-reduction services.

DRAFT recommendation(s): Existing recommendations on prisons, for sex workers and PWID on HBV vaccination

1. Prisons should have a comprehensive hepatitis programme, including the provision of free hepatitis B vaccination for all prisoners, free hepatitis A vaccination to those at risk, and other interventions to prevent, diagnose and treat hepatitis B and C equivalent to those available in the community (including condom, needle and syringe programmes and drug dependence treatment as needed).
2. Include sex workers as targets of catch-up hepatitis B immunization strategies in settings where infant immunization has not reached full coverage (Source: WHO, 2012).³⁶
3. It is suggested to offer people who inject drugs the rapid hepatitis B vaccination regimen.

Implementation considerations

• As for all testing services, programmes for key populations need to emphasize WHO’s “5 Cs” – particularly consent, confidentiality and connection to comprehensive prevention, care and treatment.
• Need to overcome reluctance to provide partner testing/index partner testing
• Make use of lay providers/peer testing for outreach especially among key populations
• Viral hepatitis testing for key populations needs to be delivered alongside other key primary prevention interventions.
• Accessibility and coverage of testing would need to be high to have an impact on the prevalence of HBV among PWID and other key populations. Offering DBS testing for HCV to PWID attending drug treatment programmes increased uptake of testing services.

Research gaps

• Further research and large scale-implementation studies should be performed to evaluate this further in other high-endemic, low-income settings.
• What proportion of HBV- or HCV-positive cases will be missed by a testing policy based on screening for at risk behaviours and exposures?
• Evaluation of different testing approaches in terms of cost, impact and cost–effectiveness and evaluation of key drivers in a range of different settings.

Topic for analysis: who to screen?

Population

1. Risk-based screening in different high-risk populations
   Injecting drug users (IDUs), men who have sex with men (MSM), immigrants, recipients of blood transfusion and blood products, sex workers, and health-care workers (HCW), HIV-infected persons
2. General population (excluding blood donors) or selected subpopulations of general population (women during pregnancy, those with raised alanine aminotransferase (ALT), infants, schoolchildren and adolescents)
3. Other approaches: Birth cohort screening (based on different age cut-offs; born between 1945–1960 or 1965 or 1970)
4. One-off screening vs repeat screening every five years.

Intervention

Testing strategies for HBV in different populations (risk-based and general population and birth cohort); and at different prevalence thresholds

Comparator

No testing or current practice or comparison of different testing strategies

Outcomes

Benefits, harms and costs, and cost-effectiveness with different screening strategies for different target populations

Individual patient outcomes: Number of cases detected, overall mortality, liver-related mortality, cirrhosis, end-stage liver disease, rate of hospitalizations, serious adverse events, quality of life

Prevention: New infections (mother to child, horizontal (IDUs needle sharing and sexual; and sexual, especially MSM)

Cost-effectiveness: Cost and incremental cost per case diagnosed; cost and incremental cost per case screened and treated; cost and incremental cost per life saved; cost and incremental cost per infections averted; quality-adjusted life-years (QALYs) gained

Background

Hepatitis C virus (HCV) is a global public health burden and major cause of morbidity and mortality including liver failure and hepatocellular carcinoma. Current global HCV seroprevalence is estimated to be 2.8%, or >185 million infected individuals worldwide.

Routes of transmission

In many countries, HBV, HCV and HIV transmission occurs predominantly in high-risk key populations, often via common routes of transmission. Key populations include people who inject drugs (PWID), people in prisons and other closed settings, some mobile populations, some indigenous populations, MSM and sex workers. PWID are a key population who are at particularly high risk of HCV infection. In many high-income countries and some developing countries, ongoing HCV transmission is driven mainly by PWID populations.

The advent of high-efficacy, low duration therapy, however, generates prioritization for testing for HCV infection, linking infected patients to care, and curing HCV before patients begin to experience the consequences of cirrhosis and end-stage liver disease.

Low rates of diagnosis and linkage to care

The majority of people are unaware of their HCV infection, and therefore often present with advanced disease. Based on still limited studies, overall <15% of the estimated 180 million who are chronically infected with HCV are aware of their diagnosis, based on data from higher-income settings - United States, Europe and China.

In addition to the very low access to and uptake of testing, there is also further attrition on the care cascade with very poor linkage to care and therefore treatment, among those who test positive.

Hepatitis B and C testing and diagnosis are at the core of entry to both the prevention and treatment cascade. Testing is required to identify those with are positive, linking them with care, counselling them on measures to reduce transmission to others then assessing who needs treatment, initiating treatment, achieving treatment response (sustained virological response [SVR] for hepatitis C). Hepatitis testing is also needed to identify those who are negative, to provide hepatitis B vaccination, and the opportunity to implement individual or facility level prevention measures counsel to reduce risk behaviours.

There are three key approaches to HCV screening

1. Targeted risk factor-based screening. This refers to screening of specific groups including key populations, who are generally at higher risk of being infected than the general population. This includes PWID, people in prisons and other closed settings, migrant populations, some indigenous populations, MSM and sex workers, but may also include health-care workers. People attending services providing care and treatment for viral hepatitis or HIV can be encouraged to bring their partners to be tested.
   This involves screening those with high-risk behaviours, exposures and other conditions. Most countries have based their list of high-risk groups as defined by the Centers for Disease Control and Prevention (CDC), and are largely based on known modes of transmission.

   1. Family members and household contacts of hepatitis B patients
   2. MSM
   3. PWID
   4. HIV-positive patients
   5. Patients on immunosuppression or chemotherapy
   6. Persons with liver disease of unclear etiology
   7. Health-care workers.
2. Birth cohort screening for HBV and HCV
3. Population or community-based screening (including antenatal). Routine general population screening: i.e. testing among the general population without attempt to identify high-risk behaviours or characteristics (“routine testing”). This means that all members of the population have access to the screening programme under consideration. It may also include home-based testing (house to house); campaigns (e.g. HTC plus - malaria, safe water, non-communicable diseases (diabetes and hypertension); outreach (mobile) in general and key populations; workplaces and schools; and health-care facility-based screening.
4. Health-care facilities. Testing could also be offered in special dedicated clinics, e.g. HIV, STI clinics. Screening at health-care facilities may include primary care settings, inpatient and outpatient settings, and may involve screening on the basis of clinical presentation or focus on only those with abnormal liver function tests, abnormal ultrasound scan, family history of liver disease or other clinical suspicion of liver function test.
   Persons in whom there is clinical suspicion of viral hepatitis: even when risk factors for HBV and/or HCV are not present, screening is indicated wherever there is clinical suspicion of viral hepatitis infection. This may occur, for example, where there is existing liver disease, including liver cirrhosis or hepatocellular carcinoma, or where there is unexplained liver disease including abnormal liver function tests.

Existing guidelines: what are countries doing?

The main approach to HCV testing is a targeted risk factor-based testing for those with high-risk behaviours, exposures and other conditions, e.g. (PWID, liver disease, renal dialysis) and use of a birth cohort approach in the US and Japan. At present, no country guidelines recommend routine testing for all individuals regardless of demographics or specific behavioural risk.

1. Screen those with high-risk behaviours, exposures and other conditions. Most countries have based their list of high-risk groups as defined by CDC, and are largely based on known modes of transmission.

   1. Family members and household contacts of hepatitis B patients
   2. MSM
   3. PWID
   4. HIV-positive patients
   5. Patients on immunosuppression or chemotherapy
   6. Persons with liver disease of unclear etiology
   7. Health-care workers.

Survey of guidelines (Surjo De)

Global epidemiology of hepatitis C virus infection: new estimates of age - specific antibody to HCV seroprevalence (Hanafiah, 2013)

Prevalence of hepatitis C antibodies in injecting drug users

A review of global prevalence data from 77 countries estimated that exposure to HCV (anti-HCV positive) among PWID is estimated to be between 60% and 80% in 25 countries and over 80% in 12 countries.

MSM

MSM can acquire HBV and HCV sexually. MSM who are HIV positive are at significantly higher risk of acquiring HCV infection than HIV negative MSM. Incidence rates of sexually acquired HCV among MSM have been rising in several industrialized countries since 2000, and outbreaks have been described in some less industrialized nations.

Systematic review of HCV prevalence in HIV-infected persons (Platt et al. 2015)

Coinfection estimates were identified for 78 of the 194 countries (40%). There were 760 HIV/HCV coinfection prevalence estimates. Findings suggest that globally HCV/HIV coinfection is 1.9% (IQR = 0·4–6·6%) among general population samples, 7% (IQR = 2·6–11·1%) among people living with HIV (PLHIV) who are pregnant or where heterosexual transmission is reported, 6.2% (IQR = 3·3–13·5%) among men who have sex with men (MSM) and 83% (IQR = 55·2–94·1%) among people who inject drugs (PWID). Odds of HCV infection are 11 times higher in the presence of HIV infection, but varied by risk group. The global estimate of HCV coinfections among PLHIV is 3.2 million (IQR = 1.4–4.3 million) of whom 1.2 million (IQR = 0.9–1.4 million) are PWID

We compared the prevalence of HCV among 105 samples of HIV-positive and -negative population groups (general population, PWID, MSM, sex workers, prison inmates and high-risk populations). This is summarized in Figure 4 and in the online table. Overall, there was a 12-fold (95% CI 11·2–11·8) increased odds for HCV positivity across all population groups among HIV-positive compared to HIV-negative persons. Odds of HCV were highest among HIV-positive prison inmates (OR=16·5 95% CI 15·9–17·1) and other high-risk populations (OR = 11·7 95% CI 11·0–12·4), PWID (OR = 5·1 95% CI 4·7–5·5), followed by MSM (OR = 3·8 95% CI 3·1–4·5) and general population samples (OR = 3·7 95% CI 3·3–4·3) and sex workers (OR = 2·5 95% CI 2·0–3·2).

Epidemic scenarios hepatitis C virus

Generally, HCV epidemics around the world are heterogeneous and represent mixtures of three core epidemic components:

1. Infection related to high-risk behaviours: In essentially every geographical region, the highest prevalence of HCV infection is among persons who use injection drugs (PWID). The prevalence of injection drug use differs between countries and regions, but within those who do inject drugs, HCV prevalence is nearly universally high. Commercial sex workers and prisoners also have increased prevalence (presumably related to both drug use and perhaps sexual transmission) as do men who have sex with men, especially those who are HIV infected. In many cohorts of PWID in North America, Europe and Asia, HCV prevalence ranges from 30% to 75%.
2. Infection related to past generalized exposures that have since been identified and removed: this epidemic pattern, in which there is a high prevalence of HCV within a given age group, is commonly referred to as a “birth cohort epidemic.” While typically identified as being the infection pattern in North America and Europe, many nations have some element of birth cohort epidemics with their unique HCV epidemiology (Table 1). Birth cohort epidemics reflect an HCV exposure source that was once present and to which a large portion of the population was exposed, but that has since been identified and removed. For example, before it was identified and sequenced, HCV infected the blood supply of many countries in all regions of the world. When the blood supply began to be screened for the presence of HCV, the exposure was removed. As a result, the incidence of HCV fell dramatically among the general population, but there remains a burden of prevalent, chronic HCV among patients who were alive and likely to get a blood transfusion during the time that HCV existed in the blood supply.
3. Generalized population epidemic: This pattern is related to a widespread exposure, often iatrogenic, that results in high prevalence (8–10%) across essentially all age groups. Note that the primary difference between a “birth cohort” pattern and a generalized pattern of infection is the duration of time that the generalized exposure existed and whether it has been removed or mitigated. An example of a generalized exposure is the common use of reusable hypodermic syringes and needles in medical settings without adequate sterilization between uses.

Few epidemics fall into one of the above three categories. Rather, most are mixed, and represent some combination of all components (Table).

Draft recommendation(s)

Existing recommendations (Source: WHO, 2014⁹)

1. Risk-based: It is recommended that HCV serology testing be offered to individuals who are part of a population with high HCV seroprevalence or who have a history of HCV risk exposure/behaviour. These include:

   • Persons who inject drugs (PWID)
   • Persons with HIV infection (HIV-positive men who have sex with men)
   • Prisoners and persons previously incarcerated
   • Persons who have had tattoos, body piercing or scarification procedures where infection control is not guaranteed
   • Children born to mothers infected with HCV
   • Close contacts of persons infected with HCV
   • Persons who have used intranasal drugs
   • Persons from a country with intermediate or high prevalence (2% or greater) of hepatitis C
   • Persons who have received medical or dental interventions in health-care settings where infection control is not guaranteed
   • Persons who have received blood transfusions prior to the time when HCV serologic testing of blood donors was initiated or in countries where HCV serologic testing of blood donations is not routinely performed.
2. General population and birth cohort: It is recommended that in settings with a high HCV seroprevalence (>8%) in the general population, testing be offered, especially at least once, to persons born between 1945 and 1955.

PWID

WHO recommends delivery of a comprehensive package of nine evidence-based interventions for HIV prevention, treatment and care for PWID, all of which are also directly relevant to prevention, treatment and care for HBV and HCV, and one of which is specific to viral hepatitis testing (vaccination, diagnosis and treatment of viral hepatitis, see Box).

Sex workers

WHO outlines a comprehensive set of interventions and approaches, both to promote enabling environments and to provide prevention, testing, care and treatment, in relation to HIV and STI programming for sex workers (see Box). These recommendations are directly relevant to the response to viral hepatitis among sex worker populations. Essential interventions include enabling sex workers to access and consistently use condoms, access prevention and care, treatment and support services, diagnosis and treatment of important comorbid conditions such as for TB (particularly in HIV endemic settings, incarcerated persons, PWID and sex workers living in exposed to poor cramped working and living conditions), other STIs, and access to harm reduction services for sex workers who inject drugs.

Importantly, as sex workers are a key population highly affected by the HBV epidemic, particularly in settings where vaccine coverage is suboptimal, and at high risk for HCV if they inject drugs, they must have access to hepatitis testing services (HepTS), repeat viral hepatitis testing, and partner and family testing wherever appropriate. The offer of HIV testing services (HTS) should be offered alongside HepTS, and HepTS can be integrated into HTS wherever possible. Delivery of HepTS should be informed by recommendations for the delivery of HTS. A variety of settings may be appropriate in which to implement testing services, including health-care settings as well as community settings, and via multiple different approaches. Outreach peer-led testing is likely to be particularly effective and acceptable in many sex worker populations.

Prisons

Key WHO recommendations around testing services for people in prisons and closed settings have, to date, mostly focused on HIV prevention, testing and treatment. In 2013, UNODC and partners developed a comprehensive package of 15 key interventions for HIV prevention and treatment in prisons and other closed settings. Due to common transmission routes, these recommendations are equally applicable to viral hepatitis, and HBV vaccination and diagnosis and treatment of viral hepatitis is one of the specific recommendations (see box). This recommendation stipulates that “prisons should have a comprehensive hepatitis programme, including the provision of free hepatitis B vaccination for all prisoners, free hepatitis A vaccination to those at risk, and other interventions to prevent, diagnose and treat hepatitis B and C equivalent to those available in the community (including condom, needle and syringe programmes and drug dependence treatment as needed).”

Relevance to different settings/populations

Will this recommendation be most relevant for particular settings (e.g. endemicity)?

Key populations

1. In all settings, a number of social and structural barriers exist to PWID being able to access testing, health-care and harm reduction services, including inadequate coverage and delivery of interventions, stigma and discrimination, high incarceration rates, unstable living conditions, comorbid health problems, poor health literacy and social difficulties. This results in inadequate uptake of prevention, testing, treatment and HBV vaccination,
2. Viral hepatitis testing and treatment for PWID must always be delivered alongside other evidence-based essential primary prevention interventions also. Studies have shown that the uptake of opioid substitution therapy (OST) and adequate needle-syringe programme (NSP) coverage, in isolation, have been shown to reduce the odds of acquiring HCV, but the combination of both interventions together had a much larger impact. Additionally, the high prevalence of comorbidities in populations of PWID, including viral hepatitis/HIV coinfection, TB, mental health problems and poly-drug use alongside social and economic predictors of poor health means that it is particularly important that comprehensive prevention, treatment, care and social services are integrated and accessible to this population.
3. In many settings, responses to viral hepatitis, with particular regard for the need to reach key populations, can be integrated in order to be most effective. When this is not possible, strong links among health services working with priority populations should be established and maintained. Additionally, integration of viral hepatitis testing and treatment with existing services for HIV diagnosis and care is likely to be effective and less resource intensive.
4. Community-based testing is a critical approach for reaching people from key populations and vulnerable populations who are unlikely to go to a facility, particularly those who are asymptomatic. To improve access to and uptake of HBV, HCV and HIV testing, community-based testing services should be made available in settings acceptable and convenient to people from key populations and vulnerable populations
5. Accessibility and coverage of viral hepatitis testing will need to be high to have an impact on the prevalence of HBV and HCV among PWID. In order to achieve this, approaches must be acceptable to PWID populations. For example, studies from the UK suggest that offering DBS testing for HCV to PWID attending substance misuse services may increase uptake of testing services.

Prisoners

Many prisons around the world have implemented viral hepatitis and HIV prevention programmes, including HepTS and HTS as part of a comprehensive package of interventions; however, they are often small in scale and lack the necessary combination of essential interventions, greatly reducing their effectiveness.

Access to testing and counselling for viral hepatitis and HIV which is voluntary in nature and easily accessible alongside access to prevention, care and treatment interventions must be urgently scaled up in prisons. Particular attention must go to providing accurate information, obtaining informed consent and maintaining confidentiality. Additionally, there are often major challenges to continuity of care within closed settings and between prisons and the community that need to be addressed.

Pregnant women

Although the risk of mother-to-child transmission (MTCT) of HCV is much lower than that of HBV, perinatal transmission of HCV does occur in 4–6% of births, and the risk is two to three times higher if the mother is coinfected with HIV. MTCT is the most common cause of HCV infection in young children. In some settings, such as in west sub-Saharan Africa, there is a relatively high prevalence of HCV among young children, which may be a result of the high prevalence of HIV/HCV coinfection in that region.

HCV risk factor information should be elicited from pregnant women, and if present, testing of pregnant women for HCV should also be considered alongside testing initiatives for HIV and HBV. There is currently no effective public health intervention to decrease the risk of MTCT of HCV. However, as DAAs become more widely available, they may have a potential role to play in preventing MTCT of HCV if found to be safe and effective for use in pregnancy. Diagnosis of women with HCV before pregnancy should be prioritized so that appropriate treatment and potential HCV clearance can be achieved.

Children

Consideration should be given to testing children born to HCV-infected mothers, particularly if the mother is coinfected with HIV or has other risk factors for HCV infection, such as injecting drug use. Similar to HBV, the progression of HCV liver disease is usually slow in infected children, but early diagnosis is important to enable monitoring for liver disease and for linkage to appropriate care and treatment when necessary. As DAAs become more widely available, HCV testing in infants and children will become important to be able to offer curative treatment at an early stage before progression of liver disease.

Most infants whose mothers have been diagnosed with HBV or HCV should be followed up and routinely offered EID, and those diagnosed with either with should be regularly monitored for signs of liver disease so that treatment can be offered when necessary. However, some infants are lost to follow up, so additional paediatric case-finding is important. This can be achieved through the routine offer of PITC in health facilities, particularly in high-prevalence settings, and also through testing the family members of index cases where appropriate.

Rationale for recommendation

Strength of recommendation

Implementation considerations

Topic for analysis: How to test

Population

Persons identified for HBV testing

Intervention

Rapid diagnostic test for HBsAg detection

Comparison

Enzyme immunoassays for HBsAg detection

Outcomes

Diagnostic accuracy (Sensitivity, Specificity, Positive predictive value, Negative predictive value, TN, TP, FN, and FP).

Background

• The most important marker for the diagnosis of hepatitis B infection that may require treatment remains the detection of hepatitis B surface antigen (HBsAg).
• Chronic hepatitis B infection is defined by the detection of HBsAg on two occasions six months apart.
• However, after initial testing, further characterization of the individual’s HBV infection is based on a sequential testing strategy of for other markers of HBV infection (supplementary testing) triggered by the detection of HBsAg in the first instance.

Immunoassays (laboratory-based)

• The most widely used HBsAg assays are laboratory-based immunoassays.
• This can be in the form of an enzyme immunoassay (EIA), chemiluminescence immunoassay (CLIA) or electrochemiluminescence immunoassay (ECL).
• These are best suited to settings with high throughput of specimens and where infrastructure (electricity, cold storage, climate-controlled rooms) and skilled staff are consistently available.
• Other simple assays such as agglutination assays are also available for detection of HBsAg but these generally require serum/plasma specimens and cold storage. The results of simple assays may be read visually.

Rapid diagnostic tests (RDTs) - performed in-laboratory or at the point-of-care

• Many laboratories in resource-limited settings may not have access to specialized equipment and process few specimens, per day. Hence, individual tests, including rapid diagnostic tests (RDTs), may be more appropriate.
• RDTs for detection of HBsAg come in immunofiltration (flow through) and immunochromatographic (lateral flow) formats.
• In general, RDTs do not require cold storage and may be tested using capillary (finger-stick) whole blood.
• The manufacturer’s instructions for use should always be followed. The results of RDTs are read visually.
• RDTs may be deliverable at the point of care (POC).
• The expansion of their use depends on their performance and operational characteristics in the setting of intended use, ultimately with the aim being to reach resource-limited settings and offer cost-efficient testing services as an alternative to assays that require specific laboratory infrastructure and staff skills to perform.

The selection of EIA or RDTs should not be mutually exclusive. Choice of appropriate technology can be complex but can usually be distilled down to three main factors: performance, cost and accessibility. There are inevitably trade-offs, based not only on disease prevalence and the health-care infrastructure, but also on technical, socioeconomic, cultural, behavioural considerations.

DRAFT recommendation(s)

Relevance to different settings/populations

Will this recommendation be most relevant for particular settings (e.g. endemicity)?

• The introduction of HBsAg testing using RDTs will be most relevant in settings where there is poor access to existing laboratory testing-services, either access to centralised laboratory testing or lack of testing-infrastructure in existing laboratories.
• Delivery of RDTs at the point-of-care in remote or resource-limited settings, e.g. HBsAg testing in antenatal clinics may significantly affect the future burden of disease.
• Useful for testing of both symptomatic and asymptomatic individuals.
• It will be most relevant to key affected populations who may be at risk of infection but who may be reluctant to or have poor access to health-care services, such as individuals who attend drug-rehabilitation clinics or prisoners. These individuals require screening, may require treatment if infected or vaccination if not currently infected.
• It will be less relevant in individuals who have good access to health care and in settings where laboratory testing for hepatitis B is already well established.

Rationale for recommendation

Strength of recommendation

Implementation considerations

• Symptomatic vs asymptomatic individuals; in a symptomatic individual, you may not need such good analytical sensitivity than when screening an asymptomatic individual.

Research gaps

• Impact of using RDTs for HBsAg at the point-of-care on delivery and implementation of testing services.

GRADE Summary of findings

1. Topic for analysis: How to test

Population

Individuals identified for HCV testing to ascertain exposure to HCV

Intervention

Rapid diagnostic tests and enzyme immunoassays for detection of antibodies to HCV

Comparison

1. Nucleic acid testing (NAT)
2. EIA and immunoblot
3. EIA only

Outcomes

Diagnostic accuracy (Sensitivity, Specificity, Positive predictive value, Negative predictive value, TN, TP, FN and FP).

2. Background

• Screening for exposure to hepatitis C virus (HCV) is dependent on assays that detect antibodies to HCV (anti-HCV) in the first instance.
• Once antibody status is confirmed, the individual should undergo supplementary testing for active HCV infection using an assay designed to detect viral replication, such as HCV RNA or core antigen (HCV cAg).
• Assays designed solely to detect antibodies to viral antigens will inevitably have a “window period” of infectivity in early infection in an individual who has been recently infected whose infection will not be detected by a given serological assay. This diagnostic window period can be shortened by direct detection of viral antigen or nucleic acid.
• The improvements in assay performance over time, in particular the EIAs, have been termed as “generations” of the assays. *(see footnote)
• It is important to note that the latest generation of assays designed to detect anti-HCV also are combined with cAg in order in increase seroconversion sensitivity of the assay, but these “4th generation” or “combination” assays should not be used to differentiate exposure from chronic infection.
• Note that sensitivity of antibody-only assays may be reduced if the patient is immunocompromised, e.g. HIV, immunosuppressive therapy, renal dialysis.

Immunoassays (laboratory-based)

• The most widely used anti-HCV assays are laboratory-based immunoassays.
• They detect antibodies to core and non-structural antigens. In the case of 4th generation assays, the assay combines detection of antibodies to HCV along with detection of hepatitis C core (P22 Ag) antigen directly.
• These can be in the form of an enzyme immunoassay (EIA), chemiluminescence immunoassay (CLIA), electrochemiluminescence immunoassay (ECL) or recombinant immunoblot assay.
• These are best suited to settings with high throughput of specimens and where infrastructure (electricity, cold storage, climate-controlled rooms) and skilled staff are consistently available.

Rapid diagnostic tests (RDTs)

• Many laboratories in resource-limited settings may not have access to this specialized equipment and process fewer specimens, per day. Hence, individual tests, including rapid diagnostic tests (RDTs), may be more appropriate.
• RDTs for detection of anti-HCV are simple to perform and do not require instrumentation, they come in immunofiltration (flow through) and immunochromatographic (lateral flow) formats and may be read visually.
• In general, RDTs do not require cold storage and may be tested using capillary (fingerstick) whole blood or oral fluid. However, the manufacturer’s instructions for use should always be followed.
• RDTs may be deliverable at the point of care (POC).
• The expansion of their use depends on their performance and operational characteristics in the setting of intended use, ultimately with the aim being to reach resource-limited settings and offer cost-efficient testing services as an alternative to assays that require specific laboratory infrastructure and staff skills to perform.

* Summary of assay “generations”

3. Draft recommendation(s)

9. Relevance to different settings/populations

Will this recommendation be most relevant for particular settings (e.g. endemicity)?

• This recommendation of the introduction of RDTs will be most relevant in settings where there is poor provision of laboratory testing-services, either access to centralized laboratory testing or lack of testing infrastructure in existing laboratories.
• Delivery of RDTs at the POC will have relevance more in remote or resource-limited settings, compared to settings where there is good access to health care and established screening programmes.
• It will be most relevant to groups of patients at risk of infection but who may be reluctant to or have poor access to health-care services, such as individuals who attend drug-rehabilitation clinics or closed settings such as prisoners. These individuals require screening and may require assessment and treatment if found to be infected.
• It will be less relevant in individuals who have good access to health care and in settings where laboratory testing for hepatitis C is already well established.

10. Rationale for recommendation

11. Strength of recommendation

12. Implementation considerations

13. Research Gaps

• Impact of testing at POC using RDTs to rule out/rule in HCV exposure on HCV-associated morbidity and mortality and onward transmission testing and treatment programmes.

GRADE Summary of findings

1. Topic for analysis: How to test

Population

Persons identified for HBV testing

Intervention

One-assay testing strategy; one HBsAg assay

Comparison

Two-assay testing strategy; two different HBsAg assays

Outcomes

Diagnostic accuracy (Sensitivity, Specificity, Positive predictive value, Negative predictive value, TN, TP, FN, and FP).

Background

• The most important marker for the diagnosis of chronic hepatitis B infection that may require treatment remains the detection of hepatitis B surface antigen (HBsAg).
• Although the case definition of chronic hepatitis B is the detection of HBsAg twice six months apart, if facilities exist, after an initial positive result, supplementary testing can be undertaken in order to facilitate entry into a care pathway.
• HBsAg will be detectable in the blood if there is current hepatitis B infection. Confirmation of the specificity of a reactive HBsAg first-line test result is usually carried out by either;

  i)

  repeating the HBsAg testing in a different assay of similar sensitivity, or

  ii)

  performing a neutralization test using a specific anti-HBs-containing reagent in the same first-line assay after appropriate dilution of the specimen under test. Specificity is confirmed when this reagent abolishes reactivity in the assay.

• This PICO question addresses the issue of whether a positive result from a single HBsAg assay has sufficient specificity in order to proceed to supplementary testing and/or entry into a care pathway, or whether confirmatory testing on the same specimen with a different HBsAg assay (or neutralization), performed sequentially after the first assay is required.

Testing strategies

• WHO recommends standardized testing strategies to maximize the accuracy of hepatitis B and C testing while minimizing cost and increasing simplicity.
• A testing strategy describes a testing sequence for a specific testing objective, taking into consideration the presumed prevalence of the analyte to be tested in the population. In both high and low prevalence settings, more than one assay may be required.
  See footnote for further detailed explanation of the one- and two-assay strategies.

Figs 1–3

Possible testing strategies for detecting HBsAg.

Since the decision for PICO 3 critically depends on whether currently available tests for detecting HBsAg show acceptable sensitivity and specificity, the following conclusions from PICO 1 should be considered:

• Overall compared to an EIA reference, RDTs had pooled clinical sensitivity 90% (95% CI: 89, 91) and clinical specificity 100% (95% CI: 99, 100).
• RDTs were more accurate among HIV-negative patients, with pooled clinical sensitivity of 93% (95% CI: 90, 95), and pooled clinical specificity of 100% (95% CI: 99, 100) compared to results in HIV-positive patients in whom RDTs are 72% sensitive (95% CI: 68, 76) and 100% specific (95% CI: 100, 100) compared to an EIA reference.
• Results for capillary whole blood specimens were comparable to serum and less heterogeneous.

The two-test strategy involves the use of a more sensitive test as the screening test and a second test that is more specific to reduce false-positive results. Since the review in PICO 1 showed that HBsAg serological assays have excellent specificity, then there is no need for the use of another test if a simple testing strategy is desired.

2. DRAFT recommendation(s)

7. Relevance to different settings/populations

Will this recommendation be most relevant for particular settings (e.g. endemicity)?

• This recommendation will be most relevant in settings that do not have established hepatitis B testing programmes and of less relevance where laboratory testing for hepatitis B is already well established.

8. Rationale for recommendation

9. Strength of recommendation

10. Implementation considerations

11. Research gaps

• Impact of a one- vs two-step testing strategy on hepatitis B testing programmes
• Cost-effectiveness studies of one- versus two-assay or neutralization testing strategies

Footnote: Explanation of one- and two-assay strategies and PPV and NPV calculation method based on prevalence

One-assay strategy

• A single test is performed. If the test result is reactive, an “HBsAg positive” status is reported, with need for complementary testing and follow-up HBsAg testing in 6 months to diagnose chronic infection recommended.
• If the initial test result is non-reactive, an “HBsAg negative” status is reported.
• This strategy efficiently identifies most uninfected individuals and rules out chronic HBV infection; it identifies those likely to be infected and in need of additional testing.
• It is a suitable testing strategy for resource-limited settings because only a single test or one assay is performed as long as an assay that meets high standards for analytical and clinical sensitivity and clinical specificity is used.
• Limitations of this approach are that a small percentage of test results may be false positive, so appropriate procedures to follow up individuals need to be in place.

Two-assay strategy

• The test results of two different assays are used sequentially (i.e. not in parallel), to increase the positive predictive value of the overall test strategy.
• If the test result on the first-line assay is non-reactive, an “HBsAg negative” result is reported. However, if the test result on the first-line assay is reactive, a second test with an assay from a different manufacturer is performed.
• If both test results are reactive, the status is reported as: “HBsAg positive” with need for supplementary testing and follow-up HBsAg testing in 6 months to diagnose chronic infection recommended.” If the test result on the second-line assay is non-reactive, the result is reported as “HBsAg inconclusive; requires additional testing.”
• This strategy efficiently identifies most uninfected individuals and more definitively rules out chronic HBV infection than a one-assay testing strategy. It improves the positive predictive value when the test results of two different assays are both reactive. It can be used by non-laboratory staff, provided that adequate quality assurance standards are in place.
• The two-assay strategy may produce a small number of false-positive results (particularly in low-prevalence settings); some persons with recent HBV infection may receive false-negative results, which will depend on the analytical sensitivity of the assays used.

Worked example to illustrate the effect of prevalence on predictive values for the two different testing strategies

Assuming the following assay performance characteristics:

• If Assay 1 has sensitivity of 99% and specificity of 98%.
• If Assay 2 has sensitivity of 99.4 and specificity of 99.5%.

Using the following equation for PPV and NPV that incorporates prevalence more correctly,

1. Topic for analysis: How to treat

Population

Persons identified for hepatitis C virus (HCV) testing

Intervention

One-assay testing strategy; one HCV serological assay

Comparison

Two-assay testing; two different HCV serological assays

Outcomes

Diagnostic accuracy (Sensitivity, specificity, positive predictive value, Negative predictive value, TN, TP, FN, and FP).

2. Background

• WHO recommends standardized testing strategies to maximize accuracy while minimizing cost and increasing simplicity.
• A testing strategy describes a testing sequence for a specific testing objective, taking into consideration the presumed prevalence of the analyte to be tested in the population. In both high- and low-prevalence settings, more than one serological assay may be required to establish exposure to HCV.
• Screening for exposure to HCV is dependent on assays that detect antibodies to HCV (anti-HCV) in the first instance. Once antibody status is confirmed, the patient will undergo supplementary testing for active HCV infection using an assay designed to detect viral replication, such as HCV RNA or core antigen (HCVcAg).
• It is important to note that the latest generation of assays designed to detect anti-HCV also are combined with cAg in order in increase sensitivity of the assay, but these “combo” assays should not be used to differentiate HCV exposure from active HCV infection.
• The question this PICO aims to address is whether one or two serological assays (anti-HCV or HCV Ag/Ab combo assays) performed sequentially are required in terms of specificity and positive predictive value in order to proceed to supplementary testing.

See footnotes for explanation of one-test and two-test strategies.

Figs 1–4

Possible testing strategies for detection of anti-HCV and diagnosis of active HCV infection.

3. Draft recommendation(s)

9. Relevance to different settings/populations

Will this recommendation be most relevant for particular settings (e.g. endemicity)?

10. Rationale for recommendation

11. Strength of recommendation

12. Implementation considerations

13. Research gaps

• Development/implementation projects evaluating use of HCV core antigen as a one-stop diagnostic test
• Impact of one-test vs two-test screening on hepatitis C testing services

GRADE Summary of findings

Footnote: Explanation of one- and two-assay strategies

Modelling tables

2. Background

• Determination of exposure to HCV through detection of antibodies to HCV (anti-HCV) is a commonly used first-line diagnostic tool to identify those who are infected with HCV who might benefit from antiviral treatment, yet diagnosis of active HCV infection requires evidence of active viral replication, traditionally ascertained by the detection of HCV ribonucleic acid (RNA) by nucleic acid testing (NAT).
• Detection of HCV RNA may also be used as the preferred first-line investigation in individuals who may be persistently seronegative due to underlying immunosuppression, e.g. poorly-controlled HIV infection, renal dialysis.
• Qualitative NAT for HCV allows for detection of the virus as well as evidence of the level of HCV RNA circulating in the peripheral blood falling below a clinically relevant threshold, i.e. a qualitative measurement.
• Assays to detect HCV-RNA have been developed for the near point-of-care (POC) setting.
• NAT assays can be very sensitive and specific but more costly than serological methods such as HCV core antigen testing, and require sophisticated laboratory equipment and therefore skilled staff.
• HCV core antigen (HCVcAg) testing was developed as an alternative to NAT for the diagnosis of active HCV infection. HCV nucleocapsid peptides 22 (p22) are released into plasma during viral assembly and can be detected throughout the course of HCV infection. There are also several assays that have been commercialized for stand-alone detection of HCVcAg as a replacement to NAT.
• Furthermore, HCVcAg is detectable earlier than antibodies and therefore detection of HCVcAg has also been applied as an additional analyte for serological assays for use to ascertain exposure to HCV as a combination HCV Ag/Ab (4th generation) assay. The addition of cAg in combination was intended to increase sensitivity of the assay in early infection. Although the output may be interrogated in order to ascertain whether antigen, antibody or both were detected, the purpose of the assays is not to differentiate seropositivity from active infection.
• Both HCVcAg and HCV RNA (either qualitative or quantitative detection) have been shown to have clinical utility for the detection of active HCV infection though there is scarce research comparing the HCVcAg, and qualitative and quantitative NAT for this purpose.

3. Draft recommendation(s)

View in own window

Summary and quality of evidence 1. Distribution of viraemia in persistent infection: A threshold of >10 000 IU/mL will capture 95% of persistent infections (except, temporarily, for a minority of those with partial viral control) between 5 and 12 months post infection. In 95% of cases of hepatitis C infections: Those with persistent infection and viral plateau will have a viral load (VL) >100 000 IU/mL at month 5 and remain at least >10 000 IU/mL between months 5 and 12. Those with persistent infection but partial viral control will also have >100 000 IU/mL at month 5 and remain having a VL at least >1000 IU/mL temporarily, going back to a viral load >100 000 IU/mL between months 10 and 12. Between 2 and 3–4 months after infection, acute infection (giving VLs up to >100 000 IU/mL) may be captured that would spontaneously resolve but would likely be antibody negative. Publications are: Hajarizadeh B, Grady B, Page K, Kim AY, McGovern BH, Cox AL et al. Patterns of hepatitis C RNA levels during acute infection: the C3 study. PLOS One. 2015;10 (4):e 0122232. Hajarizadeh B, Grebely J, Applegate T, Matthews GV, Amin J, Petoumenos K et al. J Med Virol. 2014;86 (10):1722–9. Glynn SA, Wright DJ, Kleinman SH, Hirschkorn D, Tu Y, Heldebrant C et al. Transfusion. 2005;45 (6):994–1002. 2. Systematic reviews of the evidence: Two systematic reviews were commissioned to address the above PICO questions. These aimed to summarize (1) the diagnostic accuracy of HCV cAg testing (PICO 5a), (2) the diagnostic accuracy of qualitative versus quantitative RNA for diagnosis of active HCV infection (PICO 6). Summary of results of HCV cAg review Diagnostic accuracy of HCV cAg for diagnosis of active infection (Fig. 1) 7 assays utilizing HCV core antigen were assessed. Note that two of these were HCV antigen/antibody combination assays and not designed to differentiate active infection from seropositivity, but are included as the antigen/antibody components of the assays were reported separately.							□ High □ Moderate □ Low □ Very low
			Effect Accuracy (95% CI)		Effect LR
Index test	n (samples)	Unit of analysis	Sensitivity	Specificity	Positive LR	Negative LR
Abbott ARCHITECT HCVAg Assay	20 (11,820)	Sample	93.4% (88.7, 96.2)	98.7% (96.9, 99.4)	71.8 (28.6,160.3)	0.07 (0.04, 0.12)
Ortho ELISA-Ag	5 (1,177)	Sample	93.2% (81.6, 97.7)	99.2% (87.9, 99.9)	116.5 (6.7, 977)	0.06 (0.02, 0.07)
Bio-RAD Monolisa HCV Ag-Ab ULTRA	5 (525)	Sample	28.6–95%*	94.9% (89.9, 99.8)**	NA	NA
EIKEN Lumispot HCV Ag	2 (235)	Sample	97.5–98.1%*	ND	NA	NA
Fujirebio Lumipulse Ortho HCV Ag	1 (80)	Sample	95% (90.2, 99.8)**	ND	NA	NA
Hunan Jynda HCV Core Ag ELISA	4 (524)	Sample	59.5% (46, 71.7)	82.9% (58.6, 94.3)	3.5 (1.1, 12.6)	0.28 (0.2, 0.3)
DiaSorin S.A. MurexAg/Ab EIA	4 (730)	Sample	50–100%*	83.8–100%*	NA	NA
Limits of detection of HCV cAg assays The limit of detection for the most evaluated and best Architect) is 3 fmol/L HCV cAg or 0.06 pg/mL, which equals about ~1000–3000 IU/mL with a NAT. Impact of findings in different prevalence settings (for assays where a meta-analysis was possible) (Fig. 2)
Outcome	Effect per 1000 patients with presumed HCV for varying prevalence settings comparing HCV core Ag against HCV RNA
	Prevalence 2%*		Prevalence 10%*		Prevalence 30%*
Abbott ARCHITECT HCV Ag Assay
TP	19 (18, 19)		93 (89, 96)		279 (267,288)
TN	967 (951, 974)		888 (873, 895)		691 (697, 696)
FP	13 (6, 29)		12 (5, 27)		9 (4, 21)
FN	1 (1, 2)		7 (4, 11)		21 (12, 33)
Ortho ELISA-Ag
TP	19 (16, 20)		93 (82, 98)		279 (246, 294)
TN	970 (862,980)		891 (792, 900)		693 (616, 700)
FP	10 (0,118)		9 (0,108)		7 (0, 84)
FN	1 (0,4)		7 (2,18)		21 (6, 54)
Hunan Jynda HCV Core Ag ELISA
TP	12 (9, 14)		60 (46, 72)		179 (138, 216)
TN	813 (578, 921)		747 (531, 846)		581 (413, 658)
FP	167 (59,402)		153 (54, 369)		119 (42, 287)
FN	8 (6,11)		41 (28, 54)		122 (84, 162)
Limits of detection of hepatitis C NAT assays This systematic review shows that for diagnosis of active HCV of detection of most commercial qualitative assays was in measured against a WHO standard, whereas the lower quantitative assays is at 600–1100 IU/mL. Additional information: Newer quantitative viral load assays report limits of detection similar to qualitative viral load but might not quantitate results at that level. NAT are in the pipeline for detection on capillary whole blood. Furthermore, detection from dried blood spot is also promoted to increase access. For both those strategies, the limit of detection is likely to be substantially higher (e.g. in the range of 2000 IU/mL) than for existing assays (because of volume tested and technical feasibility). Fig. 3: Limit of detection of qualitative vs quantitative NAT Lee SC, Antony A, Lee N, Leibow J, Yang JQ, Soveiro S et al. Improved version 2.0 qualitative and quantitative AMPLICOR reverse transcription-PCR tests for hepatitis C virus RNA: calibration to international units, enhanced genotype reactivity, and performance characteristics. J Clin Microbiol. 2000;38 (11):4171–9. Sarrazin C, Dragan A, Gartner BC, Forman MS, Traver S, Zeuzem S et al. Evaluation of an automated, highly sensitive, real-time PCR-based assay (COBAS Ampliprep/COBAS TaqMan) for quantification of HCV RNA. J Clin Virol. 2008;43 (2):162–8. Conclusions: HCV core Ag assays can have high sensitivity (up to 93.4% for certain commercialized assays), high specificity, and good correlation with HCV RNA to a detection limit of roughly 1000–3000 IU/mL. NAT on plasma and serum are able to achieve higher sensitivity than cAg. Qualitative assays from published data are more sensitive than quantitative assays. This might not be the case for newer assays. Quality of evidence Predictive modelling (Linas, Boston University) To be added at meeting
4. Risks/benefits cAG vs NAT Benefits HCV cAg testing by immunoassay format has the potential to be less costly given that the cost of goods is lower. Also HCV cAg is more stable and thus does not require a cold chain. Further assessment and treatment may be administered more promptly to individuals when diagnosis of active HCV infection is undertaken in a more decentralized manner. There are both HCV cAG and NAT tests in the pipeline that might be possible on capillary blood at the point of care. However, current cAg immunoassays still require sophisticated laboratory equipment, electricity and therefore skilled staff to operate, while there are already some NAT assays available that can be done near the patient (however, still require plasma). HCV cAg assay can conceivably be used in a one-step testing strategy (i.e. without an antibody test) particularly in high-prevalence settings. With such a strategy, patients can be identified earlier in their infection (than with antibody testing), results may be available faster to the patient and provider, resulting in less loss to follow up and faster treatment initiation. Risks The patient with low-level viraemia (<3000 IU/mL) could be missed with HCV cAg assays or HCV RNA NAT with lower sensitivity. The clinical implications for the individual person and on a population level are not well understood. However, if the test does reach more patients (e.g. because it can be done on capillary blood or is less costly), then this risk might be outweighed by the benefits. Although data exist on the utility of HCV cAg in seronegative individuals, no studies examined the use of HCV cAg and HCV RNA NAT in diagnosing HCV infection in key affected populations, in a one-step testing strategy rather than using an antibody assay. Such a strategy would only be cost-effective in certain high-prevalence settings. Qual versus Quant NAT Benefits Qualitative NATs generally are more or at least as sensitive than quantitative assays. The cost of a qualitative NAT assay may be lower than that of a quantitative assay. Risk Recent publications suggest that monitoring for HCV will not be necessary with direct-acting antivirals (DAA) (as viral load at EVR is not predictive of cure). Therefore, quantitation at baseline will not be necessary. This is currently not yet widely confirmed or reflected in guidelines.							□ Benefits clearly outweigh harms □ Benefits and harms are balanced □ Potential harms clearly outweigh potential benefits Are the desirable anticipated effects large? □ No □ Probably □ Uncertain □ Yes □ Varies
5. Acceptability, values and preferences A values and preferences survey of implementers and users of hepatitis B and C testing services was carried out by FIND in September 2015. A total of 104 respondents from 43 (20 high-income, 23 low- and middle-income) countries. Relating to this PICO, As assay (platform) for detection of HCV cAg is currently available in India, Indonesia, Macedonia, Viet Nam, Turkey (although the representativeness of these data are limited). The platform is available in South Africa but not currently being used for HCV cAg detection. 50% found a diagnostic sensitivity of >95% acceptable, particularly if the test cost is lower, therefore increasing access to testing (respondents felt that then the test cost should be less than US$ 10 (83%) with a sensitivity of 95%. For a test with a sensitivity of 98%, forty-one per cent of respondents would accept a cost of US$ 11–20). Free text comments suggested that HCV cAg was easier to do. A larger number of respondents felt that the cost of testing for HCV RNA by NAT was considered more of a barrier than that of HCV cAg. 47% of respondents in low- and middle-income countries favoured a decentralized test and a test on capillary blood even at the cost of sensitivity. Also, 50% of patients preferred a test result in <2 hours (which could only be achieved at the point of care) while 27% found a result on the same day acceptable. Patient: Patients at risk of progressive liver disease will benefit from reduced disease progression and related mortality and morbidity, if treatment is provided as a result of wider access to testing programmes. Community: To identify the individuals who require assessment and treatment would be an effective use of resources. As testing and treatment programmes are scaled up, the numbers developing progressive disease and serious outcomes (HCC and complications of advanced liver disease), premature morbidity and mortality within the community will be reduced, and so also the burden of disease to societies where the disease is most prevalent. Health-care workers: Appropriate use of resources to channel treatment to patients with higher risk of complications in the medium- and short term Will require training in the use of testing equipment if being used in the near-POC setting Appropriate reporting and recording of results. Laboratory: Will require training for HCV cAg test: careful sample processing is necessary for HCV cAg assay to lyse viral particles, expose antigen and dissociate antibody from antigen and optimize the detection for HCV cAg.							□ No major variability □ Major variability Is the option acceptable to key stakeholder? □ No □ Probably □ Uncertain □ Yes □ Varies
6. Equity, ethics and human right implications Will the recommendation raise questions around equity? Equity will improve as a result of decreased cost and increased decentralization of testing; however, still improvement of access to testing facilities is necessary. Regional and country variability in access to treatment. Are there ethical implications to this recommendation? Ethical consideration for the possibility that WHO could recommend a testing strategy.							□ Less equitable □ More equitable
7. Resource use and financial implications Input from modelling team Materials/equipment: Cost of testing platform and reagents Other laboratory consumables Training and supervision: Appropriate training of laboratory staff Quality control programmes If using near-POC assays, appropriate training of testing providers. Other: Cost of transportation of specimens to the laboratory Possible procurement costs (Fig. 4): Cepheid: <20 NAT. With volume-based pricing down to~15 Note that new POC or near-POC assay platforms for HCV RNA are in development. See figures at the end of the table.							Are the resources required small? □ No □ Probably □ Uncertain □ Yes □ Varies
Feasibility and constraints to implementation Are any major barriers expected for the implementation of this recommendation? None, provided there is internal and country commitment to HCV testing. Regional and country variability in access to treatment and procurement of testing equipment and services With regard to any diagnostic assay, availability of a local laboratory, which is able to procure the testing platform and reagents required for testing. Feasibility survey report to be presented at meeting.							Is the option feasible to implement? □ No □ Probably □ Uncertain □ Yes □ Varies

n: study number, CI: confidence interval; LR: likelihood ratio; ND: no data, NA: not applicable – if sensitivity and specificity results were not available from meta-analysis, likelihood ratios were not calculated.

TR: true positives (individuals with active HCV); TN: true negatives (individuals without active HCV); FP: false positives (individuals incorrectly classified as having active HCV); FN: false negatives (individuals incorrectly classified as not having active HCV)

*

Meta-analysis not possible. Range of results seen across studies reported.

**

Result from one study only.

*

Numbers in parentheses consider 95% confidence interval of accuracy estimate.

*

Refer GRADE table in footnote

8. Relevance to different settings/populations

Will this recommendation be most relevant for particular settings (e.g. endemicity)?

• Currently available HCV cAg testing will be more relevant to populations that presently rely on centralized laboratory testing for HCV RNA for confirmation of status.
• One-step strategies for testing with HCV cAg or NAT may be cost-effective only in high-prevalence settings.
• The recommendations are less likely to be relevant in high-income settings where there is already access to established hepatitis C testing and treatment programmes.

9. Rationale for recommendation

10. Strength of recommendation

11. Implementation considerations

• Optimize test for asymptomatic patients in primary-care settings or in the community where the HCV endemic is high.

12. Research gaps

• Development/implementation projects evaluating use of HCV core antigen or HCV RNA as a one-step diagnostic strategy.
• Surveillance data: how many patients are missed by assays that have limits of detection of 2000 IU.
• Outcomes of patients with low viral loads
• More information on patients with high viral loads and negative HCV cAg to inform the optimization of antigen detection.
• The kinetics of HCVcAg with treatment needs to be evaluated further, particularly in the context of new DAA regimens.
• More rigorous assessment of covariates in accuracy studies is required, such as HIV or HBV coinfection or genotype (particularly genotypes 5 and 6 where there are limited data).
• Development of muliplex instrument with other disease diagnosis with HIV, HBV, and TB at health centre.

GRADE Summary of findings

Fig. 1POC HCV platforms available within the next 2 years

Fig. 2Qualitative/Quantitative HCV RNA platforms currently available (1–5) and soon to be available (6–7)

1. Topic for analysis: How to test

Population

• Samples for serology for HBV (HBsAg)
• Samples for serology for HCV (HCV Ab)
• Samples for HBV DNA
• Samples for HCV RNA

Intervention

Using dried blood spot (DBS) samples

Comparison

Using venous samples

Outcomes

Diagnostic accuracy (Sensitivity, specificity, positive predictive value, negative predictive value, TN, TP, FN, and FP)

2. Background

In high-prevalence of hepatitis B virus (HBV) among low- and middle-income countries (LMICs) there is a need for improved HBV screening, especially in decentralized settings. And the entry of new alloral direct acting antiviral therapy for hepatitis C provides an opportunity to scale up HCV care in LMICs and dramatically simplify diagnosis and monitoring.

In HIV, use of DBS has facilitated the diagnosis of HIV in children under 18 months and is a promising tool for HIV management in resource-limited settings (viral load monitoring is strongly recommended by WHO as the preferred ART monitoring test). While commercial rapid diagnostic tests (RDTs) exist for HBsAg and HCV antibody for HBV diagnosis and HCV screening, respectively, the use of DBS sent to centralized lab facilities for diagnosis or screening purposes may be useful in certain contexts where RDTs are not available or not feasible due to human resource, procurement, quality, regulatory or other constraints, particularly as they can be prepared from capillary blood, thus obviating the need for phlebotomy services.

Molecular tests for HBV DNA and HCV RNA must currently be performed in centralized facilities, where barriers to sample collection and transport, such as phlebotomy services, plasma separation and cold chain, may make testing less feasible. One decentralized molecular platform for HCV RNA does exist but it too relies on plasma as a sample type. Thus DBS may be used in a similar way to HIV to facilitate diagnosis and monitoring in certain contexts.

Two main advantages of using DBS for HCV compared to HIV are that (1) there is no proviral DNA to overestimate the quantity of virus in the blood compared to plasma (although cell-associated RNA may still contribute) and (2) qualitative testing should suffice, both for diagnosis and measurement of SVR, as (i) viral load monitoring of DAA-based treatment is not useful and, (ii) since DAA therapy is non-toxic, a log drop calculation for treatment continuation at week 12 (EVR), as was the case for IFN-based therapy, is not needed.

In prioritizing the validation of serological testing on DBS versus NAT testing on DBS, depending on the context, they may be equally useful. If affordable, good-quality RDTs are available that can be performed off capillary blood then the impetus may be to prioritize that validation of NAT testing for HBV DNA and HCV RNA. However, if RDTs are not available then DBS testing may be equally important to increase access to serological testing, for example, for hard-to-reach populations and those with poor venous access. Equally, DBS may be useful where polyvalent screening for multiple diseases, such as HIV/HBV/HCV, is useful and where multiplex RDTs for this purpose are not available or more costly. Thus the choice and combination of test may be context specific whereby different programmes may opt for difference combinations of: (1) DBS serology + DBS NAT (remote settings), (2) RDT serology + DBS NAT (clinics, e.g. antental), or (3) EIA serology + plasma-based NAT (urban settings or more central hospitals).

See Feasibility section for current use of DBS.

3. Draft recommendation(s)

11. Rationale for recommendation

12. Strength of recommendation

13. Implementation considerations

• Regulatory approval for DBS as a sample type
• Where HR and phlebotomy services are lacking, training of lay workers to perform fingerprick DBS
• QA/QC; e-health/m-health to improve turnaround time of results
• Training clinicians to act on the result.

14. Research gaps

• Operational research to evaluate the accuracy of DBS using field specimens prepared and stored in real-life conditions will be needed.
• Validation of the use of DBS with commercially available HBsAg and HCV Ab tests and determining cut-off values with the use of ROC curves will be needed.
• The use of DBS for HCV RNA viral load measurement and of the rate of degradation of HCV RNA when stored in DBS at ambient temperatures and high hum idity for different time periods will be needed as well as further study in HIV-coinfected patients and for use in treatment monitoring, specifically for measuring SVR 12/24 post DAA therapy.
• A systematic review on the LOD for HCV RNA testing on DBS that will serve to capture (1) everyone with chronic and active HCV infection, and (2) everyone failing DAA therapy at SVR12/24 (which will then inform the best time-point for SVR).
• A systematic review on the LOD for HBV DNA testing on DBS that will serve to capture (1) everyone who needs HBV treatment, and (2) everyone failing therapy.

GRADE Summary of findings for PICO7a

GRADE Summary of findings for PICO7b

GRADE Summary of Findings for PICO7c

GRADE Summary of findings for PICO7d

Population

Patients receiving treatment for HCV

Intervention

HCV core antigen assay

Comparison

NAT for HCV RNA detection (and/or) quantification

Outcomes

Diagnostic accuracy (Sensitivity and specificity, TN, TP, FN, and FP)

1. Background

• HCV core antigen (HCVcAg) testing was developed as an alternative to NAT for diagnosis of active HCV infection. HCV nucleocapsid peptides 22 (p22) are released into plasma during viral assembly and can be detected throughout the course of HCV infection.
• Detection of HCV viraemia is also important during treatment of chronic HCV infection.
• Current guidelines recommend verification of virological activity pre-treatment with the measurement of a baseline HCV RNA quantitative measurement (viral load) by NAT. For interferon-based treatments, HCV RNA viral load is assessed at week 4 of therapy for the “rapid viral response” (RVR) to help predict efficacy of therapy, and repeated at week 6 if elevated at week 4 to see further viral response and guide whether treatment should be continued.
• NAT for HCV RNA is performed again at week 12 (early viral response, EVR), at the end of treatment, and 12 and 24 weeks after therapy is completed to test for cure, “sustained viral response” (SVR).
• New, direct-acting antivirals (DAA) have made treatment for HCV much easier with oral rather than parenteral administration and shorter, more effective regimens that are likely to be easier to adhere to making access to affordable diagnostic and monitoring assays even more important. However, it is important to note that, ultimately, treatment monitoring may not be required with the routine use of DAAs.
• This PICO addresses the question of whether HCVcAg can be used as a tool for assessing response to treatment for HCV infection.

2. Draft recommendation(s)

8. Relevance to different settings/populations

Will this recommendation be most relevant for particular settings (e.g. endemicity)?

• HCVcAg testing will be more relevant to populations that presently rely on centralized laboratory testing for HCV RNA for confirmation of status.
• The recommendations are less likely to be relevant in high-income settings where there is already access to established hepatitis C testing and treatment programmes.

9. Rationale for recommendation

10. Strength of recommendation

11. Implementation considerations

Optimize test for asymptomatic patients in primary-care settings or in the community where the HCV endemic is high.

12. Research gaps

• The kinetics of HCVcAg with treatment needs to be evaluated further, particularly in the context of new DAA regimens.
• More rigorous assessment of covariates is required in studies assessing HCVcAg or NATs for treatment monitoring or as a test of cure, such as HIV or HBV coinfection or genotype.
• Is treatment monitoring and/or confirmation of cure necessary with DAA regimens? If so, what would be the optimal timing of testing?
• When is the best time-point to test for cure with HCV core Ag?
• Development of muliplex instrument with other disease diagnosis such as HIV, HBV, and TB at health centre.
• HCVcAg assay to detect the variants of HCV.

GRADE Summary of findings