the care of HIV–HCV coinfected patients changed significantly in “real life” in France between 2004 and 2009

In conclusion, the care of HIV–HCV coinfected patients changed significantly in “real life” in France between 2004 and 2009. These results underline the importance of continuing efforts to educate physicians and patients about increasing the access of coinfected patients to HCV treatment. Efforts should be made to improve the care of these patients, such as increasing the access to and the number of patients receiving treatment for HCV infection, the possibility of starting HCV treatment earlier (i.e. before the presence of a high liver fibrosis score), the introduction of HCV treatment even in patients with low CD4 cell counts, and earlier referral of patients with cirrhosis to a liver transplant unit.....RESULTS: The rate of liver biopsy decreased (14% vs. 38% vs. 56%), while the use of non-invasive liver damage tests increased (47% vs. 24% vs. ND) in the 2009, 2006, and 2004 surveys, respectively. The rate of patients that had never been treated for HCV infection progressively decreased in the 2009, 2006, and 2004 surveys (37%, 42%, and 54%). The main reasons for HCV non-treatment changed as HCV treatment was deemed less questionable and the lack of liver biopsy became a very rare reason (6%, 18%, and 34%). Among patients treated for HCV infection, the rate of sustained virological response increased (49%, 29%, and 29%). The main factors independently associated with HCV treatment were a liver fibrosis score >F2 (odds ratio=3.5; 95% CI 2.1–5.7), a liver biopsy activity grade A2 (2.7; 1.4–5.3), a CD4 cell count 350ml (2.7; 1.6–4.4), European origin (2.1; 1.3–3.4), daily alcohol consumption<30g (2.1; 1.2–3.8), and male gender (2.0; 1.2–3.3)."


Articles in Press


Care of hepatitis C virus infection in human immunodeficiency virus-infected patients: Modifications in three consecutive large surveys between 2004 and 2009


Journal of Hepatology June 2010

Patrice Cacoub12, Philippe Halfon3, Eric Rosenthal4, François Bailly5, Firouze Bani Sadr6, Yves Benhamou7, Stéphane Chevaliez8, Jean Michel Pawlotsky8, Lionel Piroth9, Yazdan Yazdanpanah10, Stanislas Pol11


Received 30 November 2009; received in revised form 9 February 2010; accepted 15 March 2010. published online 05 May 2010.

Corrected Proof

ABSTRACT

Background & Aims


To analyze the care of HCV infection in HIV–HCV coinfected patients and its progression between 2004 and 2009.

Patients and methods


Three hundred eighty HIV–HCV coinfected patients were prospectively included from November 22 to 29, 2004 (2004 survey), 416 patients from April 3 to 10, 2006 (2006 survey), and 419 patients from June 15 to 22, 2009 (2009 survey).

Results


The rate of liver biopsy decreased (14% vs. 38% vs. 56%), while the use of non-invasive liver damage tests increased (47% vs. 24% vs. ND) in the 2009, 2006, and 2004 surveys, respectively. The rate of patients that had never been treated for HCV infection progressively decreased in the 2009, 2006, and 2004 surveys (37%, 42%, and 54%). The main reasons for HCV non-treatment changed as HCV treatment was deemed less questionable and the lack of liver biopsy became a very rare reason (6%, 18%, and 34%). Among patients treated for HCV infection, the rate of sustained virological response increased (49%, 29%, and 29%). The main factors independently associated with HCV treatment were a liver fibrosis score >F2 (odds ratio=3.5; 95% CI 2.1–5.7), a liver biopsy activity grade A2 (2.7; 1.4–5.3), a CD4 cell count 350ml (2.7; 1.6–4.4), European origin (2.1; 1.3–3.4), daily alcohol consumption<30g (2.1; 1.2–3.8), and male gender (2.0; 1.2–3.3).

Conclusion


Compared to the 2004 and 2006 surveys, the 2009 coinfected patients had liver damage assessment more frequently, more patients had received HCV treatment and more patients had achieved a sustained virological response.

Introduction


Coinfection with hepatitis C virus (HCV) and human immunodeficiency virus (HIV) is a frequent and particularly serious problem [1], [2], [3], [4], [5], [6]. About 30% of HIV-infected patients in France are also infected with HCV, i.e. close to 30,000 patients [1], [2], [5], [6]. Since the widespread use of combined antiretroviral therapy (cART) was begun, AIDS mortality has progressively decreased, while chronic liver disease, linked primarily to HCV, has become one of the leading causes of morbidity and mortality [2], [4], [6].


Considerable therapeutic progress has been achieved in HCV-infected patients due to combination therapy with pegylated interferon and ribavirin, resulting in a 55–60% sustained virological response (SVR) in HCV mono-infected patients (40–50% in HCV genotype 1, the most frequent HCV genotype worldwide) [7]. Results from large therapeutic trials in HIV–HCV coinfected patients have provided important information. A particularly encouraging result was that the pegylated interferon and ribavirin combination achieved an SVR in 27–44% of coinfected patients [8], [9], [10], [11], resulting in histopathological improvement in most of them. Recent results report similar SVR rates in coinfected more than in HCV mono-infected patients [12], [13], [14], [15]. HCV treatment, however, is given to a small number of HIV–HCV coinfected patients for many reasons, e.g., HCV treatment deemed questionable (minimal hepatic lesions, alcohol abuse, active drug use), lack of available liver biopsy, psychiatric contraindication, and physicians’ conviction of poor patient compliance [16], [17], [18], [19], [20], [21], [22], [23], [24], [25]. Following the First European Consensus Conference on the treatment of chronic hepatitis B and C in HIV-infected patients in 2005, the coinfection guidelines in HIV medicine from the European AIDS Society in 2008 as well as their recent update at the European AIDS Clinical Society conference, it may be assumed that differences in the guidelines may have had an impact on treatment compliance and management standards [26], [27].


The aim of the present study was to analyze the care of HIV–HCV coinfected patients in France in 2009 (including initial workup, mode of treatment, follow-up modalities, and virological response) through a survey and to compare results with surveys conducted in 2004 and 2006 [17], [22].

Patients and methods


The same methodology was used in the three surveys, and details have been published elsewhere [17], [22]. In brief, physicians involved in the management of HIV-infected patients were recruited from 50 specialized centers from all of metropolitan France. These centers were representative of those providing the standard of care in HIV cART in France. Each physician was asked to prospectively fill out a standardized data collection form for all HIV–HCV coinfected patients seen between November 22 and 29, 2004 (2004 survey), April 3 and 10, 2006 (2006 survey) and June 15 and 22, 2009 (2009 survey). The form included sociodemographic data of the patient and physician, the HIV and HCV virological status, information regarding the pre-therapeutic workup of HCV infection, the type of HCV treatment, and its follow-up. In the 2006 and 2009 surveys, some questions were added to obtain more details on current addiction (alcohol consumption, IVDU, and substitution), liver damage assessment (biopsy or non-invasive tests), modalities of the follow-up of patients with cirrhosis, HCV treatment modalities and their virological results [sustained virologic response (SVR) was defined as an undetectable HCV-RNA (<50IU/ml) sixmonths after the end of HCV treatment], use of erythropoietin, use of anti-depressants, and the type of HIV treatment. This form had been pre-tested by four physician specialists managing HIV–HCV coinfected patients in order to optimize the type and means of data collection.

Discussion


In the last decade, chronic liver disease, linked primarily to HCV, has become one of the leading causes of morbidity and mortality in HIV-infected patients [2], [4], [6]. In the present study we had the opportunity to analyze the care of HIV–HCV coinfected patients in France in 2009 (including initial workup, mode of treatment, follow-up modalities, and virological response) and to compare the results with those obtained in the 2004 and 2006 surveys. We showed that the care of HIV–HCV coinfected patients has changed significantly in “real life”: (1) patients had more liver damage assessment; (2) more patients had received HCV treatment; and (3) more patients achieved a sustained virological response.


The complete workup of HCV infection in coinfected patients has progressed dramatically, mostly due to better liver damage assessment. In the 2009 and 2006 surveys, non-invasive liver tests had been done in 72% vs. 45% of patients using serum biomarkers (52% vs. 67%), transient elastography (63% vs. 11%), or both (25% vs. 22%). Numerous groups in France have been leaders in the development of non-invasive techniques for liver fibrosis assessment and this may partially explain the increased use in our surveys. The absence of liver biopsy was given as one of the main reasons for non-treatment in the 2004 survey, while this reason was given in only a small percentage of patients in the 2006 and 2009 surveys. Fibrometer, Hepascore, and FibroTest were shown to be able to stage liver fibrosis in HIV–HCV coinfected patients with area-under-the receiver operating characteristic curves (AUROCs) of 0.89, 0.84, and 0.78 for the diagnosis of >F2, respectively [28]. In 114 HIV–HCV coinfected patients with liver biopsy performed before and sixmonths after the end of HCV treatment, non-invasive biological scores showed a significant decrease in SVR patients [29]. Evaluations of liver stiffness using transient elastography have shown cut-off values from 4.5 to 7.2kPa for the diagnosis of significant fibrosis, with AUROCs from 0.72 to 0.83; performances were even better for the diagnosis of cirrhosis with AUROCs from 0.95 to 0.97 [30], [31]. The dual combination approach of serum biomarkers and transient elastography for the diagnosis of significant fibrosis could avoid the use of liver biopsy. Other methods have also been developed, including a novel panel of blood markers [32] and a genomics-based approach [33].


The percentage of coinfected patients who had never received HCV treatment decreased from 54% to 37% between 2004 and 2009, while the number of reasons cited by physicians for non-treatment decreased from 2.4 to 1.8 (out of the 11 choices presented on the data collection form). The reasons for non-treatment were numerous and their profile also changed [34], [35], [36], [37], [38], [39]. HCV treatment was deemed less questionable, the absence of liver biopsy no longer seemed to be a major barrier, and the number of contraindications to HCV treatment decreased (mainly due to fewer psychiatric reasons). Barriers to HCV treatment were more frequent in HIV–HCV coinfected than in HCV mono-infected patients, mainly due to depression and ongoing alcohol and injection drug use [19], [24], [40]. Fleming et al. [16] reported a cohort of 149 HIV–HCV coinfected patients, 70% of whom were ineligible for HCV treatment mainly due to non-compliance with medical visits, drug or alcohol use, and active psychiatric disease. Maida et al. [36] reported that only 40.1% of 405 coinfected patients had already been exposed to interferon-based therapies. Other groups found even smaller proportions (<2–33%) of coinfected patients who had undergone HCV treatment [24], [37], [38]. More recent data from the EuroSIDA group however showed results quite similar to those found in the present study. In European countries, the incidence of starting HCV treatment in HIV–HCV coinfected patients increased from 3.9 per 1000 patients prior to 1998 to 32.6 per 1000 patients in the year follow-up after 2004. In multivariate analysis, an estimated 23% annual increase in the incidence of starting HCV treatment was found when adjusting for baseline factors [39]. Programs are successfully integrating hepatitis C care for IVDUs into health-care settings [20]. Studies in methadone maintenance patients showed rates of SVR in 28–94% of patients with excellent completion rates of HCV treatment (72–100%) [41]. Many aspects of the barriers to HCV treatment can be overcome with direct treatment, such as the anti-depressant drugs used in only 20% and 22% of patients in the 2006 and 2009 surveys, respectively [21], [41], [42], [43]. Asthenia may be a consequence of treatment-related anemia during HCV therapy [44], [45]. Erythropoietin treatment should then be considered to maximize SVR in HCV therapy; it had been used in 7% of patients in the 2006 survey and 34% in the 2009 survey. Its use may help to maintain full doses of standard of care treatment, especially ribavirin, by improving the quality of life during the treatment [46], [47].


A dramatic increase in the SVR rates in coinfected patients between 2004 and 2009 was seen in the present study. In the 2004 survey, we did not have specific data available concerning the duration and type of HCV treatment. In the 2009 and 2006 surveys; however, most treated patients (94% vs. 80%, respectively) had received the combination of pegylated interferon and ribavirin. It is striking to note that an SVR was obtained in 29% of patients in the 2004 and 2006 surveys and in 49% of those in the 2009 survey. This figure is very close to that reported (up to 54%) in recently published large randomized trials [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [47]. Based on data from large therapeutic trials and clinical studies (reflecting a more pragmatic attitude) [48], about half of HIV–HCV coinfected patients receiving HCV treatment will have an SVR. In terms of coinfection and its particular seriousness [49], [50], [51], [52], these results seem particularly encouraging [27].


In HIV–HCV coinfected patients with end-stage liver disease, liver transplantation shows promise as a potential therapeutic option. In one study [53], 2- and 5-year survival rates after liver transplantation were 73% and 51%, and 91% and 81% in 35 coinfected and 44 mono-infected patients, respectively. In another study [54], survival at 1, 2, and 3years was 83%, 75%, and 62% in the 12 coinfected vs. 98%, 89%, and 84% in the 59 mono-infected patients, respectively. However, more recent data showed the 5-year survival rate to be significantly lower in coinfected vs. mono-infected subjects, with liver complications secondary to HCV recurrence as the main driver of this poor outcome [55]. Thus, it does not seem that transplantation may be the final solution for most decompensated cirrhotic patients with HIV–HCV. Earlier referral of these patients to a liver transplant unit after the first episode of liver decompensation (which was rarely the case in the two more recent surveys), the use of new drugs that are effective against HCV, and the avoidance of drug toxicity (i.e. cART-related hepatotoxicity) are mandatory for improving the results of this challenging indication for liver transplantation.


In the present study, the immuno-virological control of HIV infection from 2004 to 2009 showed increased rates of undetectable HIV viral load (63–79%) and CD4 level counts >350/ml (49–71%). This control was even better in the sub-group of patients who received HCV treatment [25]. Pegylated interferon and ribavirin responses tended to increase with higher CD4 counts with similar rates of side effects [56], although there are conflicting results [57]. High CD4 cell counts, the lack of markers of severe liver disease and therapy against HCV were factors associated with a better hepatic outcome in HIV–HCV coinfected patients treated with cART [58], [59]. However, more data are needed in order to better define the impact of different cART on SVR rates after HCV treatment in such patients [60], [61].


In conclusion, the care of HIV–HCV coinfected patients changed significantly in “real life” in France between 2004 and 2009. These results underline the importance of continuing efforts to educate physicians and patients about increasing the access of coinfected patients to HCV treatment. Efforts should be made to improve the care of these patients, such as increasing the access to and the number of patients receiving treatment for HCV infection, the possibility of starting HCV treatment earlier (i.e. before the presence of a high liver fibrosis score), the introduction of HCV treatment even in patients with low CD4 cell counts, and earlier referral of patients with cirrhosis to a liver transplant unit.

Results


Sixty-two, 58, and 71 physicians following HIV–HCV coinfected patients participated in the 2009, 2006, and 2004 studies, respectively (two-thirds of physicians participated in all surveys). The profiles of these physicians did not differ in the three surveys, as they practiced in university hospitals (70–74%), general hospitals (22–25%), or other areas (3–7%). They were from departments of infectious disease (52% vs. 47% vs. 39%, respectively), internal medicine (33% vs. 26% vs. 27%), HIV/AIDS information and care (12% vs. 7% vs. 17%), hematology (0% vs. 3% vs. 10%), hepatogastroenterology (3% vs. 7% vs. 6%) or other areas (15% vs. 10% vs. 1%).


During the weeks the studies were carried out, 419, 416, and 380 HIV–HCV coinfected patients were included in the 2009, 2006, and 2004 surveys, respectively (Table 1). Individual patients did not undergo repeated assessments in the different surveys. They were mainly male (about 71% in the three surveys) with a mean age of 45.1±7.3 vs. 43.5±6.3 vs. 41.6±6.6years (36–50years in 80% of patients). Most patients in the three surveys were of European origin (73–82%). Non-European patients were mostly of African origin, i.e. 22% vs. 17% vs. 21% in the 2009, 2006, and 2004 surveys, respectively. In the 2009 and 2006 surveys (data not available in the 2004 survey), 35% of patients had excessive alcohol consumption: 53% vs. 53% with <30g/d, 32% vs. 28% with 30–50g/d, and 15% vs. 13% with >50g/d, respectively; 82/378 (22%) and 70/410 (17%) patients were or had been IVDUs, 72 vs. 56 patients of whom received an opiate substitution treatment.

Patients had been diagnosed with HIV infection for 12.8 vs. 13.7 vs. 12.1years (from 6 to 20years in 68%, 84%, and 77% of patients, respectively). The therapeutic management of HIV infection appeared to be very appropriate as evidenced by high rates of undetectable HIV viral load (<50copies/ml) and CD4 levels >500/ml, and a last measured HIV viral load done an average of 1.8 vs. 2.6 vs. 2.1months before the date when the survey was conducted (87%, 83%, and 91% of patients, respectively, had had an HIV viral load assessment done in the previous threemonths) (Table 1). In 2009 and 2006 (data not available in the 2004 survey), 381/419 (91%) and 377/416 (91%) of the patients received combination anti-retroviral therapy(cART) for HIV, i.e. boosted protease inhibitor based (61% vs. 66%), non-nucleoside based (23% vs. 18%), triple combination of nucleosides (3% vs. 2%), or other regimen (13% vs. 14%).


Patients had been diagnosed with HCV infection for an average of 9.5 vs. 9.6 vs. 8.5years in 2009, 2006, and 2004, respectively (from 1 to 5years in 18%, 13%, and 18%; 6–15years in 52%, 71%, and 70%; and more than 15years in 21%, 12%, and 6%) (Table 1). Patients in 2009 compared to those in 2006 and 2004 had a similar distribution of HCV genotypes (genotype 1 or 4 in 68%, 65%, and 65%, respectively). The extent of the pre-therapeutic HCV workup depended in part on the year of the survey. The liver damage assessment in particular significantly changed over time. The rate of liver biopsy alone decreased 14% vs. 38% vs. 56%, while the assessment of liver damage by means of non-invasive markers only increased (47% vs. 24%, data not available in 2004). There was no correlation between the performance of a liver biopsy and physician specialization. In the 2009 and 2006 surveys (data not available in 2004), non-invasive liver tests had been performed in 72% vs. 45% of patients using serum biomarkers in 172/330 (52%) vs. 124/185 (67%), transient elastography in 224/358 (63%) vs. 20/185 (11%), or both in 100/398 (25.1%) vs. 41/185 (22%) patients. Overall, the proportion of patients who had a liver biopsy or non-invasive markers has increased (86% vs. 82% vs. 56%). A complete workup, including HCV-RNA, HCV genotyping and a liver damage assessment by biopsy or non-invasive tests was performed in 58% vs. 63% vs. 47% in the 2009, 2006, and 2004 surveys, respectively (p<0.001).


The analysis of HCV treatment showed that 165/398 (42%) vs. 199/393 (51%) vs. 100/380 (26%) patients had been previously treated (p<0.01), whereas 21% vs. 16% vs. 17% of patients were receiving ongoing treatment in the 2009, 2006, and 2004 surveys, respectively (Table 2). The main reasons for non-treatment of HCV in 131 (42%) vs. 164 (42%) vs. 205 (54%) patients in the 2009, 2006, and 2004 surveys changed (Table 2), while the number of reasons per patient decreased (1.8 vs. 2.1 vs. 2.4). When HCV treatment was deemed questionable, the reasons given were: minimal hepatic lesions for 53% vs. 47% vs. 32% of patients, chronic excessive alcohol consumption for 35% vs. 29% vs. 31% and active drug use for 20% vs. 6% vs. 12%. The main contraindication to HCV treatment was a psychiatric disorder (83% vs. 67% vs. 73%), while all other contraindications were noted in 5–26% of patients (cardiovascular, poor general condition, neurological, immunological, hematological, etc.). Physician conviction of poor compliance with HCV treatment was centered on the reluctance of the patient (54% vs. 70% vs. 48%) and unfavorable socioeconomic conditions (43% vs. 33% vs. 31%). Patient refusal increased from 16% to 27% of patients. There was no correlation between the number of treated patients and HCV genotype distribution in any of the surveys. Among treated patients, an SVR to HCV treatment was noted in 81/165 (49%) vs. 50/167 (29%) vs. 27/92 (29%) patients; there was an absence of virological response in 40 (24%) vs. 73 (43%) vs. 40 (43%) patients; and a relapse or virological breakthrough in 44 (27%) vs. 44 (26%) vs. 25 (27%) patients, respectively. SVR rates were not significantly correlated to HCV genotypes. Genotype 1, 4, 5 or 6 vs. genotype 2 or 3 showed SVR rates of 53% and 47% vs. 41% and 59% vs. 44% and 56% in the 2009, 2006, and 2004 surveys, respectively. If the optimal duration of HCV treatment in coinfected patients is considered to be about 12months, 31% vs. 53% vs. 58% of patients had stopped their combination therapy prematurely. In the 2009 and 2006 surveys (data not available for the 2004 survey), most treated patients received a combination of pegylated interferon plus ribavirin, 94% vs. 80%. Among these, 22% vs. 20%, were also receiving an anti-depressant drug, and 34% vs. 7% were receiving erythropoietin treatment. In the 2009 and 2006 surveys, cirrhotic patients (128 and 83 patients, respectively) had a follow-up including liver ultrasound exam (mean interval of 7.1 vs. 7.5months) and upper GI endoscopy (mean interval of 13.9 vs. 10.2months), whereas only 10% vs. 8% had been referred to a liver transplant unit.

Compared to non-treated patients in all surveys (Table 3), patients who had received HCV treatment were more likely to be of European origin, had better control of HIV infection (more frequent undetectable HIV viral load, less frequent low CD4 cell count), and were more likely to have had a liver damage assessment. In a multivariate analysis in the 2009 survey (Table 4), the main characteristics that were independently associated with HCV treatment were a liver fibrosis score >F2 (on biopsy, biomarkers or transient elastography), a liver biopsy activity grade >A2, a CD4 cell count >350ml, European origin, daily alcohol consumption <30g, and male gender.