Intravenous Drug Use in HCV Infection

From Journal of Viral Hepatitis

Intravenous Drug Use

Not a Barrier to Achieving a Sustained Virological Response in HCV Infection

H. Jafferbhoy; M.H. Miller; J.K. Dunbar; J. Tait; S. Mcleod; J.F. Dillon

Posted: 02/21/2012; J Viral Hepat. 2012;19(2):112-119. © 2012 Blackwell Publishing

Abstract and Introduction

Abstract

Hepatitis C virus (HCV) is commonly transmitted by intravenous drug use (IDU) but drug users are under represented in many treatment cohorts, this is because of the assumption of lowered treatment success. We assessed HCV treatment outcomes in active intravenous drug users and patients on opiate substitution therapy. The Tayside HCV treatment database was retrospectively analysed for consecutively treated patients based on risk factor for acquisition of HCV. Primary end point was sustained virological response (SVR). Two hundred and ninety-one consecutively treated patients were assessed. The overall SVR rate was 55.3%. The SVR rates by risk factor were; Non-IDU 61.4%, Ex-IDU 54.8% and Active IDU 47.1% (P = n/s). In the groups G1 patients SVR was; Non-IDU 52.7%, Ex–IDU 30.7% and active IDU 35.4% (P = n/s). In the non-G1 patients: non-IDU 65.1%, Ex-IDU 76.7% and active IDU 53.5%. Ex-IDU had a significantly better SVR than active IDU, other differences were not significant. Our results demonstrate that SVR rates in the active drug users and those on opiate substitution therapy can be achieved which are comparable with non-IDU infected individuals. Intravenous drug use in those engaged with treatment services should not be seen as a barrier to treatment of HCV.

Introduction

Hepatitis C (hepatitis C virus, HCV) is a chronic blood borne viral infection. According to the World Health Organization, 2.2% or 170 million people in the world are infected and there is wide geographical variations.^[1,2] Today, the cardinal risk factor in the Western World is intravenous drug use (IDU).^[3] Despite various prevention strategies, the incidence of HCV in the IDU population has remained high ranging between 30% and 90%. Currently the standard treatment is a combination of pegylated interferon and ribavarin which, depending on the HCV genotype and disease stage, can lead to cure rates ranging between 30% and 80%.^[4–6] The key variable in treatment success is adherence to the arduous regime, often up to 48 weeks duration.^[5]

The need for adherence to therapy and monitoring, both to ensure patient safety and successful treatment, was often used to regard IDU as a relative contraindication to therapy.^[7,8] This was both on grounds of safety and the fear that lowered treatment success would reduce the cost effectiveness of therapy. However the barriers to treatment in the HCV population with coexisting drug misuse are multifactorial.^[9] The patient, care provider and the system all need to be streamlined in a fashion to provide equal treatment opportunity to this vulnerable group. Key to this is collaboration between HCV specialist services and drug staff to avoid fragmentation of services, a greater awareness of the risks posed by HCV to these individuals in the long term and the benefits of therapy, with such care pathways in place it may increase adherence with therapy in this group.

Aim

To ascertain, in routine clinical practice, the outcomes of treating individuals with HCV who are active intravenous drug users or are on opiate substitution therapy such as methadone. The primary outcome measure was the rate of sustained virological response (SVR) in those from an IDU background compared with those infected by other aetiologies. We hypothesize that a patient focussed treatment pathway negates other influences on achieving a SVR in this group of patients.

Methods and Patient Profile

HCV Population

The NHS Tayside Health Board in the East of Scotland serves a population of approximately 400 000. We know that 1012 individuals are alive and still resident in this area and have tested positive for Anti-HCV antibodies. Of these 926 are also HCV RNA positive and of those, 721 have attended HCV specialist services either in hospital or in out reach clinics on at least one occasion. The treatment cohort is all the patients who have been treated with pegylated interferon from this accurate general population-based group and is representative of the diagnosed patients within Tayside. For all individuals in the cohort, we reviewed medical records; data collected included demographical information, data on risk factors, laboratory tests, follow-up and treatment. All patients in the study were allocated a code, and patient information used was anonymous.^[10]

Care Pathway

The care pathway for patients evolved during the period of observation. It started as a conventional primary care to secondary care referral, with patients reviewed in a consultant-led outpatient service. Post 2004 it evolved to a service where drugs workers, prison nurses and other workers directly involved in care of high risk patients could make referrals, to a rapid nurse led service that was located in hospital and community settings, including drug problems services and prisons. Under the new scheme most patients were seen within 6 weeks of referral, patients referred by nonmedical practitioners (e.g. drug workers) were often seen in a locality close to the patient and often with the worker who had referred them who would have an ongoing generic support role during therapy. This resulted in a statistically significant increased number of patients who accessed care. If individuals did not attend their first appointment, they were not automatically reappointed; however, a letter was sent to the individual, and their referrer. Appointments were rebooked if the individual or referrer contacted the service at any point. The ongoing care of the patient and their therapy was administered in the nearest outreach clinic to the patients home, co-located with other ongoing therapy services, e.g. addiction treatment services or conventional hospital clinics, according to patient preference. This complemented with excellent communication between all the health care professionals and other key workers involved with the patient, who would have opportunity to support adherence to therapy. To meet the increase demands to this service 0.5 of a whole time equivalent nurse was added to the service.^[10]

Therapy

During this period individuals were treated with pegylated interferon either α2a 180 μg/week or α2b 1.5 μg/kg/week and ribavarin given by weight. The treatment length was either 24 weeks if genotype 2/3 or 48 weeks if genotype 1 and 4 with the exception of human immunodeficiency virus (HIV) co infected patients who were treated for 48 weeks irrespective of genotype. In genotype 1 patients, if a 12-week early virological response (2 log drop in viral load) was not achieved, therapy was stopped.

Treatment was also stopped on patient request or if they developed severe cases of cytopenia, uncontrolled worsening of psychiatric conditions or decompensated liver disease, this was defined as incomplete therapy

Patient Groups

The patient's aetiological risk factor for HCV infection was self-defined. Based on the risk factor for HCV transmission the patients were divided into three categories based on information in the database and case notes. Group 1 acquired infection without acknowledged intravenous use of drugs. This Group comprised of patients who had haemophilia, blood and product transfusion, body piercing and tattooing, sexual transmission identified as route of transmission. Those with IDU as a risk factor were divided into two groups; study group two active drug user and group three former IDU. Group 2 included patients who admitted that they had used intravenous drugs within the last year though the frequency was not recorded on the database this included many patients on opiate substitution therapy. We did not have access to any urine toxicological testing for those patients on substitution therapy. For the purpose of analysis all patients on methadone were included in this grouping of active drug use, as this group are dependent on opiates and illicit drug use is very common and there was no way of confirming the absence of intermittent illicit drug use. Use of nonintravenous drugs of abuse in this group was common, but we do not have reportable data for this. This cohort represent the 'real world' patients who are engaged enough with services to accept referral and who are being considered for anti-HCV therapy. This group will include a range of patients with differing degrees of stability in their lives, but there are no universally accepted criteria for sub-dividing this grouping further, that we could apply retrospectively.

Former IDU group (Group 3), defined as patients who had not used intravenous drugs in the past 12 months and were not on substitution therapy. This cohort also included individuals who started treatment during incarceration; this totalled 11 patients for the period of the study. Patients were not excluded if they had co-infection (hepatitis B virus/HIV) or did not complete therapy in an effort to reflect real life clinical practice.

Data Collection and Analysis

All HCV treatment data is prospectively entered into the Tayside portion of the Scottish Hepatitis C clinical database. We retrospectively analysed our Hepatitis C treatment outcomes and identified all patients who had been commenced on pegylated interferon therapy. The primary outcome of treatment was SVR and this is presented as an intention to treat (ITT) and a per protocol analysis (PPA). Patients who did not achieve an SVR were divided into: Nonresponder – a patient who after antiviral therapy for HCV had detectable HCV RNA at the end of treatment, Relasper – a patient who had no detectable HCV RNA at the end of treatment, but who had detectable HCV RNA 6 months after the end of antiviral therapy.

Results are presented as mean and standard deviations. Statistical analysis was performed with the Student t-test for continuous parametric variables, chi squared for discontinuous variables and Fishers tests were done for non-parametric variables. A multi-variate analysis was performed. Binary logistic regression was used to determine the adjusted odds ratios of achieving sustained virological response in illegal drug users given history of drug use, demographic variables, and genotype. Because the model was developed to determine the adjusted odds ratio rather than make accurate predictions all variables of interest were forced into the model using the Enter Method in the statistical program SPSS version 15 (SPSS Inc., Chicago, IL, USA). The accuracy of the model is measured by both the Hosmer–Lemeshow goodness-of-fit statistic and the area under the receiver operating characteristic curve produced by the predictions of the regression model. Statistical probability at the 5% level was considered statistically significant.

Results

In total 291 HCV patients were consecutively treated with combination therapy of pegylated interferon and ribavarin. Of these patients 98% were treatment naïve. The cohort included patients co-infected with HIV (10), hepatitis B (12) and haemophilia (19).

Nondrug Users (Non-IDU)

This group comprised 122 patients with an average age of 43.89 (±10.71) years, a body mass index (BMI) of 26.27 (±5.02) and a slight male predominance 76 (62.3%).39 (31.9%) genotyping revealed genotype 3 was most common: 71 (58.2%) followed by G1 36 (29.5%). Cirrhosis was present in at least 16 patients, 13.1% of this cohort. In pretreatment blood results the mean alanine aminotransferase (ALT) was >2 times the upper limit of normal at 92.39 U/L (85.4) and the mean GGT was 1.5 times the upper limit of normal 87.63 U/L (69.48).

Intravenous Drug Users

The IDU group had 169 individuals in whom the average age was 39.33 (8.39) years with a BMI of 24.29 (3.57). Predominant gender in this group was male 130 (76%). Major Genotypes were G1 71 (41.5%) and G3 86 (50.3%). Cirrhosis was seen in 16 (9.3%). The pretreatment blood test revealed that ALT and GGT were 1.5 times the upper limit of normal being 64.80 U/L (86.86) and 83.01 U/L (88.31) respectively.

Active Drug User

The active drug users were 87 in total with an average age of 36.78 (7.54) and a BMI of 24.02 (3.13) with male predominance 64 (72.7%). 29(33.3%) admitted to having recently injected drugs. The rest were on methadone but no urine toxicology data is available to refute that they were not actively using drugs simultaneously. Interestingly there were less genotype 1 patients, 31 (35.22%) than G3 47 (53.40%). Cirrhosis was seen in a small percentage (3.4%, 3 patients).

Former Intravenous Drug User (Ex-IDU)

There were 82 patients in this category with an average age of 42.04 (8.45) and BMI of 24.52 (3.91). Of this group 65 (78.3%) were male and there was an equal split of genotype 1 and 3 (46.98% and 46.98%). 13 (14.7%) had evidence of cirrhosis.

The patient characteristics of drug users (ex-IDU and active DU combined) and nondrug users are shown in Table 1. Patients without drug use as a risk factor were older, there were more women and they were more likely to be genotype 3. Patients who had a history of IDU had a lower BMI of 24.29(3.57), P = 0.0013 compared with non-IDU patients. Additionally, genotype 1 infection was seen more commonly in the IDU population (P = 0.04). Biochemically, the only significant difference was ALT, non-IDU patients had a higher ALT level pretreatment than the IDU group (P = 0.0005). All biochemical values were pretreatment values.

The patient characteristics of active and ex-IDUs are demonstrated in Table 2. In comparison between the ex-IDU and active DU the latter group was significantly younger, ex-IDUs had a significantly higher proportion of patients with cirrhosis. The major genotypes 1 and 3 were not significantly different between the two groups.

Sustained Virological Response

It was intended to treat the whole cohort of 291 patients with the aim of eradicating their HCV infection; therefore we have performed an ITT analysis on this whole cohort. However 14 of the patients in this cohort completed therapy successfully but did not return for SVR testing; in the ITT cohort they have been assumed to have failed to have achieved an SVR, thus giving this cohort a worse case scenario. In reality a proportion of this group have achieved an SVR, to illustrate this we have performed a PPA, where these 14 have been excluded but all other patients in the cohort are analysed including those whose therapy was stopped for medical reasons or by patient request.

SVR: Intention to Treat Analysis

The overall SVR rate in the entire cohort of 291 including the 14 (4.8%) individuals who moved out of the area before an SVR could be measured was 55.3%. The SVR rates achieved, by sub-group, were; 61.4% in non-IDU, 54.8% in ex-IDU and 47.1% in active DU, this was not statistically different. The SVR for genotype 1 patients in each category were 52.7% (19/36) in non-IDU, 30.7% (12/39) in ex-IDU and 35.4% (11/31) in active DU, this was not statistically different. The SVRs for the nongenotype 1 patients in each of the three groups were 66.1% (56/86) in non-IDU, 76.7% (33/43) in ex-IDU and 53.5% (30/56), in active DU. Ex-IDU patients had a significantly higher SVR than active DU patients. There was no significant difference compared with non-IDU, as shown in Table 3 and Fig. 1. Table 4 shows total number of patients and those who did not achieve an SVR in each group. This is broken down by genotype to show; the relapsers, nonresponders, patients with incomplete treatment due to medical reasons and patient requests to stop therapy. There was a higher response rate in the active IDU group but this was balanced by an increase in patient terminated therapy in this group.

Figure 1. Shows percentage of drop outs and sustained virological responses (SVR) based on intention to treat (ITT) and per protocol analysis (PPA) seen in the three groups of patients.

Sustained Virological Response Per Protocol Analysis

This analysis was performed on the cohort of 277 patients, 14 patients were excluded (4 non-IDU, 2 ex-IDU and 8 active DU) from the analysis as no SVR data was available. In all cases this was due to the patient moving out with the area prior to SVR measurement. This resulted in improvement of overall SVR rate to 58.1%. Similarly SVR rates achieved in the non-IDU patients was 63.5% with genotype 1 achieving a cure rate of 54.2% and non-genotype 1, 67.4%. The ex-IDU patients achieved an overall SVR of 56.2%, with genotype 1 attaining 32.4% and non-genotype 1 remaining at 76.7%. In the active DU group overall SVR was 51.2%, for genotype 1 42.3% and non-genotype 1 56.6%. There was no statistically significant difference between any of the IDU sub-groups compared with the non-IDU group (Table 5 and Fig. 1).

Logistic Regression Model

The logistic regression model included 267 cases. Those without a 6-month follow up of their initial treatment response were excluded along with missing SVR results. So this was performed as an ITT analysis. The independent variables for the model were genotype, age, sex, injecting behaviour and methadone prescription. There was no significant difference between the adjusted odds ratios of achieving sustained virological resistance in non-IDU compared with the ex-IDU and current IDUs. However age and genotype were still strongly predictive of successful SVR (Table 6).

Discussion

Hepatitis C is a major health risk and economic burden, which could be alleviated by the delivery of therapy and cure to those infected. In Tayside there were just under 1000 individuals diagnosed with HCV, at the end of the study period: of those, over 70% have attended at least one appointment at the specialist treatment services and of those 40% have been successfully commenced on treatment. This is considerably higher than the pyramid of care reported by National Institute of Clinical Excellence and other surveys.^[11] Clearly, this proportionately large unselected treatment cohort will accurately reflect real world issues and include those with ongoing issues around drug misuse. Indeed given the relative ease with which this part of the population at risk of HCV infection can be identified they are frequently over represented in unselected cohorts. If treatment for HCV infection is to make a difference to the natural history of the HCV epidemic at a population level then those infected with HCV must be brought into therapy with no bias against the IDU population. Therapy for HCV is regarded as cost-effective,^[12–14] but this evaluation is critically dependent on the success of the therapy, the SVR, and this is highly dependent on adherence to therapy. This concern over lack of adherence reducing the efficacy of therapy has been the traditional reason for exclusion of those patients with an IDU background from therapy. This study seeks to answer this question: is the concern over the treatment of IDU patients with chronic hepatitis C justified on the basis of SVR?

We have assessed the treatment outcome in 291 consecutively treated, predominately treatment-naive patients who received pegylated interferon and ribavarin for HCV. In contrast to much of the UK, due to record linkage, we know this represents 31.4% of the total diagnosed population in our community. They represent a typical unbiased clinic cohort. Of the three groups the non-IDU cohort was older and slightly overweight. As expected they had a higher number of cirrhotic individuals, though this does achieve significance when compared to the IDU group (Ex and active).

The sustained virological response was the highest in the non-IDU group with 61.5%, followed by 54.8% in the ex-IDU patients, with active IDU patients achieving an SVR of 47.1% of patients. This represents the worse case scenario as those patients with missing SVR data were regarded as treatment failures. This outcome was particularly common in the active IDU group, which contained a much higher number of prisoners, who upon liberation were lost to our care, as they did not reside in Tayside. These patients had fully completed therapy and could be expected to have high SVR rates. We therefore performed an analysis excluding these missing values. Unsurprisingly this improved the SVR rates for all groups and narrowed differences between the groups, which continued to show no significant differences other than ex and active IDU with nongenotype 1 infection. The overall SVR rates are slightly lower than the landmark randomized control trials (RCTs) however none of the trials treated active drug users and ex-IDU were only included if stable for >12 months.^[5,6] Secondly our non-G1 patients are almost exclusively genotype 3, which will have a worse outcome compared to that of the non-genotype 1 RCT groups which have a large genotype 2 cohort. It has been shown in various cost effectiveness analysis that combination therapy prolongs life, improves quality of life and is cost-effective.^[15] But these analyses were based on the results of the RCTs as models and hence excluded patients with intravenous drug use. These studies are based on patients with chronic HCV achieving overall SVRs of 55% (95% CI 52–58%) and for genotype 1 SVRs of 36–46% and for nongenotype 1 of 61–76%.^[4–6] The results of our investigation fall within this range with the exception of the ex-IDU genotype 1 subgroup, which had an excess of cirrhotic patients, however the reported values in this study are still within the range to make the therapy cost effective. Clearly genotype is the most significant predictor of SVR. When this is taken into account the reassuring theme that comes across is whether it is genotype 1 with the longer treatment course where the SVR rate in the active drug user was 42.3%, or the relatively shorter treatment of non-G1, where the active IDU had treatment success in 56.6%, these results lie in the SVR range that has been demonstrated to be cost-effective. The regression analysis performed found the usual predictors of SVR such as genotype and age, but did not find significant reductions in chance of SVR in those from IDU backgrounds, while the confidence intervals in the analysis are wide due to sample size, the sample did identify genotype and age as factors reducing the chance of SVR and we still offer therapy to older genotype 1 patients. Overall the results do not differ significantly in the non-IDU group, thus confirming the message: active IDU is not a barrier to treatment or a successful achievement of SVR.

Another key question is: What is the additional patient support needed to achieve the SVRs reported in this paper? A local redesign of HCV treatment services introduced a new referral pathway in 2004, implementing strategies to improve attendance and outreach clinics within the Tayside Drug Problems Service. This model utilizes existing staff working smarter in an extended role, minimising additional staffing costs. This led to an increase in referral rates. These changes in services generated a large number of patients in the early stages of drug addiction treatment coming forward for assessment of HCV. This provided us the opportunity to assess efficacy of therapy in such a group of patients.^[10] In essence, the pathway is dependent upon nurse specialists delivering the anti-viral therapy; this would now be regarded as the standard of care for HCV therapy. The care pathway is patient focussed, so for patients in active treatment for addiction the HCV therapy is co-delivered with their addiction therapy, thus minimizing the need for the patient to travel to other clinics and intimately involving the drug key worker as well as the nurse specialist in the patient's treatment. This involves the nurse specialist in outreach work but requires little additional expenditure on staff other than minimal transport costs, particularly as several patients can be seen in one session.

Our results do show that with simple support provided by a dedicated team of nurses, drug addiction workers, social services and physicians working in partnership can result in SVR rates in the active drug user group comparable with non-IDU infected individuals. This maintains the cost effectiveness of antiviral therapy for HCV in these patient groups. The intervention is designed around existing staff working smarter, rather than expensive alternatives hence a generalizable solution for the National Health Service and possibly throughout the world. Finally, it offers the patient the prospect of a better chance of cure and a positive healthcare experience.

References

1. The Global Burden of Hepatitis CWG. Global burden of disease (GBD) for hepatitis C. J Clin Pharmacol 2004; 44(1): 20–29.
2. Shepard CWFL, Alter MJ. Global epidemiology of hepatitis C virus infection. Lancet Infect Dis 2005; 5(9): 558–567.
3. Hahn JA, Page-Shafer KLP, Bourgois PSE et al. Hepatitis C virus seroconversion among young injection drug users: relationships and risks. J InfectDis 2002; 186(11): 1558–1564.
4. Fried MW, Shiffman ML, Reddy KR et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med, 2002; 347(13): 975–982.
5. Manns MP, McHutchison JG, Gordon SC et al. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet 2001; 358(9286): 958–965.
6. Hadziyannis SJ, Sette H, Morgan TR et al. Peginterferon-Î ± 2a and ribavirin combination therapy in chronic hepatitis C. Ann Int Med 2004; 140(5): 346–355.
7. Edlin BR, Seal KH, Lorvick J et al. Is it justifiable to withhold treatment for hepatitis C from illicit-drug users? NEngl J Med 2001; 345(3): 211–214.
8. Reimer JSB, Castells X, Schafer I et al. Guidelines for the treatment of hepatitis C virus infection in injection drug users: status quo in the European Union countries. Clin Infect Dis 2005; 15(40): S373–S378.
9. Morrill JA, Shrestha M, RW G. Barriers to the treatment of hepatitis C. Patient, provider, and system factors. J Gen Intern Med 2005; 20(8): 754–758.
10. Tait JM, McIntyre PG, McLeod S, Nathwani D, Dillon JF. The impact of a managed care network on attendance, follow-up and treatment at a hepatitis C specialist centre. J ViralHepatitis. 2010; 17(10): 698–704.
11. Parkes J, Roderick P, Bennett-Lloyd B, Rosenberg W. Variation in hepatitis C services may lead to inequity of heath-care provision: a survey of the organisation and delivery of services in the United Kingdom. BMCPublic Health. 2006; 6: 3.
12. Bernfort L, Sennfält K, Reichard O. Cost-effectiveness of peginterferon alfa-2b in combination with ribavirin as initial treatment for chronic hepatitis C in Sweden. Scand J Infect Dis 2006; 38(6-7): 497–505.
13. Buti M, Medina M, Casado MA, Wong JB, Fosbrook L, Esteban R. A cost-effectiveness analysis of peginterferon alfa-2b plus ribavirin for the treatment of naive patients with chronic hepatitis C. Aliment PharmacolTherapeut 2003; 17(5): 687–694.
14. Shepherd JBH, Cave CB, Waugh NR, Price A, Gabbay J. Clinical- and cost-effectiveness of pegylated interferon alfa in the treatment of chronic hepatitis C: a systematic review and economic evaluation. Int J TechnolAssess Health Care 2005; 21(1): 47–54.
15. Sroczynski G, Esteban E, Conrads-Frank A et al. Long-term effectiveness and cost-effectiveness of antiviral treatment in hepatitis C. J Viral Hepatitis 2010; 17(1): 34–50.

Source