Individuals infected by HIV exhibit high survival rates when they have access to effective combination antiretroviral therapy (cART).1 The immune recovery associated with cART has decreased mortality from AIDS-related opportunistic infections or malignancies.2 However, these individuals remain at increased risk of death from other conditions (liver-related diseases, notably).3 Infection by the hepatitis B virus (HBV) or the hepatitis C virus (HCV) and excessive alcohol consumption are also highly prevalent in this population.3–5 Furthermore, cART can be hepatotoxic in its own right but may also induce hepatotoxicity indirectly by impairing lipid and glucose metabolism (eg, by triggering the development of insulin resistance).6,7
In the HIV-positive (HIV+) population, liver-related mortality is mainly due to end-stage liver disease.4,8 There is recent evidence to suggest that death from hepatocellular carcinoma (HCC) is increasing4,8 and that this trend is likely to persist (despite improvements in antiviral treatments).9–11 However, reliable data on HCC management in HIV+ patients are scarce.9 This may be because of the persistent controversy over the influence of HIV infection on the natural history and prognosis of HCC.12–14 Whereas the benefit of routine screening for HCC has been demonstrated in HIV-negative (HIV−) cirrhotic patients, this issue has not been investigated in their HIV+ counterparts. Hence, the present study set out to analyze and compare treatment access, delivery, and outcome in HIV+ patients with HCC and matched HIV− controls.
PATIENTS AND METHODS
Selection of the Population and Diagnostic Criteria
Since January 2004, all HCC patients treated in our university medical center have been prospectively evaluated in a dedicated, multidisciplinary, oncology staff meeting. In fact, our hospital is a tertiary referral center for hepatobiliary diseases and collaborates closely with affiliated hospitals, one of which has a Department of Infectious Diseases with a marked interest in HIV disease.15,16 Between January 2004 and December 2009, 473 consecutive treatment-naive patients with HCC were evaluated in our center and formed the basis of the present study population.
HCC was diagnosed in strict compliance with the criteria issued by the European Association for the Study of the Liver17 and the American Association for the Study of Liver Disease (AASLD),18 whereas HCC staging was performed according to the Barcelona Clinic Liver Cancer (BCLC) staging system.19 Chronic liver disease and cirrhosis were defined by a Metavir fibrosis score of ≥3 or 4, respectively. Finally, HIV staging was based on the 1993 Centers for Disease Control and Prevention (CDC) criteria.20
Procedures for the Dedicated Multidisciplinary Team Meeting
Our multidisciplinary team meeting (MDT) procedures have been described elsewhere in detail.21 In brief, the MDT was attended by physicians from different disciplines and always included (in addition to the referring physician) a liver surgeon, a gastroenterologist, an oncologist, a radiologist, and a specialist nurse, at the very least. All specialists were senior and the same specialists were present throughout the study period. Our operational model includes 3 major components: (1) a weekly meeting, in which only HCC cases are discussed, (2) the full range of therapeutic options available on-site (liver transplantation, resection, ablation, chemoembolization, and systemic drug therapy, including angiogenesis inhibitors), and (3) a specific outpatient visit. Whenever a treatment is suggested by the MDT, the patient then undergoes a specific outpatient visit with a surgeon, an interventional radiologist or an oncologist (as appropriate), during which the treatment is explained and any potential limitations for performing this treatment are reassessed (eg, previously undiagnosed comorbidities, tumor progression, deteriorating liver function, technical difficulties and contraindications, etc). If limitations are identified, the case is reevaluated by the MDT and an alternative treatment option is chosen. The discussion processes and conclusions for each patient were recorded prospectively.
The treatment of HCC was classified as being curative or palliative, according to international guidelines.22 The former class includes surgical resection, radiofrequency ablation, and liver transplantation, whereas the latter class includes transarterial chemoembolization, systemic chemotherapy, angiogenesis inhibitors, and the best possible supportive care.18 After treatment, all patients were monitored at 3- to 4-month intervals; the follow-up included a clinical examination, standard clinical biochemical assays [including serum alpha-fetoprotein (AFP) levels] and imaging [computed tomography (CT) or magnetic resonance imaging, rather than ultrasonography].
Study Design and Endpoints
After approval by our institutional review board, a 3-step approach was initiated and executed. We first compared the clinical and tumor-related characteristics of HIV+ and HIV− patients. Next, treatment access and outcomes in HIV+ patients were compared with those in a matched group of HIV− patients. Matching was performed in a 1:2 ratio by gender, etiology of liver disease (viral hepatitis, alcoholic, nonviral chronic liver disease, nonalcoholic chronic liver disease), and disease severity [Child–Turcotte–Pugh (CTP) scores], tumor characteristics (BCLC stage and vascular invasion), and the treatment year (±1 year). Finally, we used a cohort approach to enable evaluation of the whole cohort in a survival analysis, while controlling for confounding factors.
The primary endpoint was overall survival, defined as the time interval between the date of initiation of HCC treatment and the date of patient death (irrespective of the latter's cause). For the purposes of the present study, survival was updated in February 2012. Secondary end points were (1) the treatment suggested by the MDT, (2) the treatment actually delivered, and (3) survival stratified by the patient's treatment type (curative or palliative) and by the patient's degree of participation in pretreatment HCC screening programs. Recommended screening for HCC included abdominal imaging (ultrasound or contrast-enhanced CT or magnetic resonance imaging) and a serum AFP assay, as recommended by the 2005 AASLD guidelines.18 Compliance with the HCC screening programs was considered to be effective only when the patients had actually completed the required investigations (liver function tests, the serum AFP assay, and imaging studies) and had attended the outpatient clinic at 6-month intervals.23
Continuous variables are described as the mean ± SD or the median (interquartile range) and categorical variables are described as a percentage. Categorical data were compared by using a χ2 test or Fischer exact test, whereas continuous variables were compared by using a t test or the Mann–Whitney U test, as appropriate. Survival curves were built according to the Kaplan–Meier method and compared in a log rank test. Multivariate Cox proportional hazard regression analysis was used to identify variables influencing survival. All analyses were carried out using R statistical software (version 2.14.0, The R Foundation, http://www.r-project.org/). The threshold for statistical significance was set to P < 0.05.
Characteristics of the Study Population in General and the HIV+ Patients in Particular
Between 2004 and 2009, we evaluated 473 new HCC patients, 450 of whom were HIV− and 23 (4.9%) of whom were HIV+. Of the 23 HIV+ patients, 15 (65%) patients were referred by the Department of Infectious Diseases. The time interval between HIV diagnosis and HCC diagnosis was 1.4 years (range, 0–2 years). At the time of HCC diagnosis, 22 (95%) patients were on cART, 21 (91%) had an undetectable HIV viral load, 20 (87%) had CD4 cell counts more than 200/mm3 and 13 were CDC stage A (56%).
The characteristics of the HIV− and HIV+ patients are shown in Table 1. The HIV+ patients were significantly younger (by about 10 years) and were more likely to be seropositive for HCV, HBV or both HCV, and HBV. Other epidemiological features (including the presence of cirrhosis (in 95% of subjects), the CTP stage, and the results of liver function tests) were similar in the 2 groups. On average, the HIV+ patients were at a less advanced BCLC stage than the HIV− patients, although this difference was not statistically significant (P = 0.07). The 2 groups did not differ significantly in terms of serum AFP levels and the extent of vascular invasion (as assessed from imaging data).
Management of HIV+ Patients and their Matched Controls
Although the control group of HIV− patients was matched with HIV+ patients for some prognostic variables (etiology, severity of the underlying liver disease, the BCLC stage and tumor extension, and the year of diagnosis) only, the 2 groups were also similar in terms of other features (notably the circumstances of HCC diagnosis and the serum AFP concentrations) (Table 2). In contrast, the age (which was not a matching criterion) was significantly more in the HIV− group than in the HIV+ group (59 and 49 years, respectively; P = 0.002).
Importantly, the treatments suggested by the MDT were the same for HIV+ and HIV− patients, with curative treatment recommended in 74% of cases in both groups (Table 3). However, the 2 groups differed in terms of the curative treatments recommended by the MDT but not subsequently performed (with patients receiving palliative treatment instead). This situation concerned 30% (7/23) of the HIV+ patients and 15% (8/46) of the HIV− patients. Nevertheless, this difference was not statistically significant (P = 0.23). The main reasons for not implementing curative treatment were poor visualization of the tumor precluding ablation (n = 1 and 2 in the HIV+ and HIV− groups, respectively), rapid tumor growth (n = 2 and 4, respectively), underestimation of the patient's general or hepatic condition (n = 3 and 1, respectively), and patient refusal (n = 1 and 1, respectively). These reasons did not seem to depend on the patient's HIV status.
Survival of HIV+ Patients and their Matched HIV Controls
For the study population as a whole (n = 69), the 1- and 3-year overall survival rates were 72% and 46.5%, respectively. The HIV+ and HIV− patients did not differ significantly in terms of the 1-year survival rate (65% vs. 76%, respectively) or the 3-year survival rate (44% vs. 48%, respectively) [hazard ratio (HR) = 0.64 (0.3–1.3); P = 0.2] (Fig. 1A). Likewise, the survival rates for HIV+ and HIV− patients did not differ significantly after stratification for the treatment actually performed (data not shown).
In 16 of the 69 cases, HCC had been diagnosed outside a regular screening program. Poor compliance with attendance at screening programs was associated with impaired liver function (CTP class B: 31% vs. 7.5% in noncompliant and compliant patients, respectively; P = 0.026); CTP class C, 25% vs. 7.5%, respectively; P = 0.077), a more advanced tumor stage (BCLC C: 19% vs. 3.8%, respectively, P = 0.077; BCLC D: 25% vs. 9%, respectively, P = 0.197), larger nodules (mean ± SD size: 62 ± 52 mm vs. 34 ± 23 mm, respectively, P = 0.036] and poorer survival (Fig. 1B). The proportion of compliant patients was lower in the HIV+ group than in the matched HIV− group (61% vs. 85%, respectively), although this difference was not statistically significant (P = 0.09). The 3-year survival rate was 15% in noncompliant HIV+ patients and 61% in compliant HIV+ patients [HR = 5.8 (1.7–20), P = 0.0056] (Fig. 1C). Curative treatment was performed in 22% (2/9) of the noncompliant HIV+ patients, 57% (8/14) of the compliant HIV+ patients, and 56% (26/46) of the HIV− patients.
In a multivariate analysis performed in HIV+ patients and their matched HIV− controls, compliance with HCC screening programs [HR = 0.16 (0.06–0.45), P = 0.00043], CTP stage C [HR = 38.5 (2.3–625.6), P = 0.01], and the size of the largest nodule [HR = 1.04 (1.01–1.06), P = 0.01] emerged as independent prognostic factors for overall survival. Importantly, HIV status was not significantly (P = 0.37) and independently associated with survival.
Influence of HIV Status on Survival in the Entire Cohort
In patients who received curative treatment (146 of the 473 cases), the 1-, 3- and 5-year survival were 92% [confidence interval (CI): 85% to 95%], 80% (CI: 72% to 87%), and 72% (CI 65% to 84%), respectively. Serum AFP and the number of nodules were independently associated with survival. HIV status was not an independent fa ctor for survival after adjusting for other confounding factors (Table 4).
In this analysis of a cohort of patients with HCC, HIV+ status did not have a significant impact on access to treatment or survival. The influence of HIV seropositivity on prognosis in HCC has been addressed in 3 other studies–all of which were retrospective noncontrolled trials with very long (>10 years) inclusion periods.12–14 The trials yielded conflicting results. First, Puoti et al14 examined the Italian Cooperative Group on AIDS and Tumors registry, extracted a series of 41 HIV+ patients, and compared the latter with 2 unrelated cohorts of HIV− patients. The median survival time in the HIV+ group (6 months) was significantly shorter than in the HIV− group (18 months). However, 60% of HIV+ patients had not received any treatment or had only received supportive care. Second, Brau et al13 identified 63 HIV+ patients managed at 16 centers in North American and compared them with HIV− controls observed in 4 of the centers. The respective survival rates in HIV+ and HIV− patients were similar, although the HIV+ patients had been treated more recently and more aggressively, on average. Finally, Berretta et al12 studied 104 HIV+ and 484 HIV− patients from 4 centers in Italy. The survival rate of HIV+ patients was lower than that of HIV− patients—even though a significantly higher proportion of HIV+ patients had received treatment. The median survival time in HIV− patients was also surprisingly high (59 months), although only 33% had received potentially curative treatment. To avoid the drawbacks of previous study designs, we analyzed a consecutive series of HIV+ and HIV− patients with HCC, managed by the same standardized conditions, in the same institution, and over a relatively short period of time.
Our study results confirmed previous findings whereby HCC occurs at a younger age in HIV+ patients, on average (at a mean age of 49 years in our HIV+ patients, which compares well with the values of 42–52 reported in previous studies 12–14). A background of HBV or HCV coinfection was almost always present, as also recently emphasized in the literature.12,24,25 Furthermore, 95% of our HIV+ patients were on cART—the proportion observed in HIV− infected patients who experienced liver-related mortality in the 2005 French Mortavic study8 and only slightly higher than the values observed in a multicenter series of HIV+ patients with HCC treated in the 2000s12 and in the 2003 Mortavic study.4
In the literature, the extent of active treatment of HCC in HIV+ patients is highly variable because between 30% and 60% of patients are not offered treatment or are offered treatments whose efficacy is unproven.12–14 By using a matched control group of HIV− patients to account for differences in prognostic variables, we showed that HIV+ and HIV− patients can gain equal access to HCC treatments. Curative treatment was suggested by the MDT in 74% of the patients; this high value reflects the relatively early tumor stage and our policy of strictly following international guidelines (which do not depend on HIV status).
Although we found that initially recommended curative treatment was not delivered in 30% of HIV+ patients after further evaluation (compared with just 15% of HIV− patients), this difference was not statistically significant. Furthermore, curative treatment was withheld for the same reasons in HIV+ and HIV− patients; these included the inability to perform ablation (because of technical problems, such as the tumor's topography or poor visualization),26 rapid tumor progression, impaired general or hepatic health status condition, and refusal by the patient (as described above). Nevertheless, almost half of the HIV+ patients ultimately underwent curative treatment. This proportion is similar to that recently reported by Berretta et al12 and does not differ significantly from the value found for HIV− controls.
In the present study, survival of HIV+ patients and of their matched HIV− control was not significantly different. Tumor characteristics and compliance with screening program, not HIV serology had a significant influence on outcome. Likewise, HIV serology did not independently influence survival after curative treatment in the entire cohort. This observation is consistent with the recent report whereby HIV infection does not influence overall and recurrence-free survival in HCC patients after liver transplantation.25
HIV+ patients who were not compliant with HCC screening programs had a less favorable outcome than compliant HIV+ patients. This may be because of the fact that noncompliant patients were more likely to have a worse CTP status and BCLC stage. Furthermore, there was also a trend toward less frequent diagnosis of HCC in organized screening programs for HIV+ patients. HCC screening and patient education in this population should therefore be improved and should occur earlier in the course of the disease. These compliance characteristics are not specific to HIV carriers27,28 but several factors in this population can contribute to lower adherence to prescribed screening programs, such as complex medication regimens and the need to address 2 different serious diseases (ie, HIV infection and liver diseases). This may require more focused information, personalized care, and recall policies.23
Our study had a number of limitations. First, the study population was quite small but nevertheless corresponded to the largest single center study reported to date. Second, our study's follow-up period was relatively short, so we cannot rule out the possibility that the risk of long-term recurrence is greater in HIV+ patients. However, the shape of the actuarial survival curves in both the present study and previous studies12,25 argues against this possibility. Third, the proportion of our patients with undetectable HIV viral load and a CD4 count >200/mm3 was 91% and 87%, respectively. These values are slightly higher than those reported previously.12–14,24 This, we believe, is the consequence of effective current medical treatment of HIV rather than selection bias. Nevertheless, our conclusions may not be applicable to HIV+ patients with low CD4 counts or failure to maintain undetectable viral load during ART, as both features are risk factors for end-stage liver disease.29
In conclusion, HIV+ status does not have a significant impact on access to treatment or survival in a standardized care management system. In terms of public health, HIV+ subjects may therefore benefit from active screening for HCC as much as HIV− patients do and should not be denied the opportunity of curative treatment.
All authors are rightly credited with authorship; Conception and design: O. Farges, A. Gervais, and F. Degos; Acquisition of data: C. Lim, N. Goutte, and M.-P. Vullierme; Analysis and interpretation of data: O. Farges, N. Goutte, A. Gervais, and D.C. Valla; Drafting of the article: O. Farges, C. Lim, N. Goutte, and M.-P. Vullierme; Critical revision: A. Gervais, D.C. Valla, and F. Degos; and Final approval: C. Lim, N. Goutte, A. Gervais, M.-P. Vullierme, D.C. Valla, F. Degos, and O. Farges.
1. Bhaskaran K, Hamouda O, Sannes M, et al.. Changes in the risk of death after HIV seroconversion compared with mortality in the general population. JAMA. 2008;300:51–59.
2. Simard EP, Engels EA. Cancer as a cause of death among people with AIDS in the United States. Clin Infect Dis. 2010;51:957–962.
3. Weber R, Sabin CA, Friis-Moller N, et al.. Liver-related deaths in persons infected with the human immunodeficiency virus: the D: A: D study. Arch Intern Med. 2006;166:1632–1641.
4. Rosenthal E, Pialoux G, Bernard N, et al.. Liver-related mortality in human-immunodeficiency-virus-infected patients between 1995 and 2003 in the French GERMIVIC Joint Study Group Network (MORTAVIC 2003 Study). J Viral Hepat. 2007;14:183–188.
5. Salmon-Ceron D, Lewden C, Morlat P, et al.. Liver disease as a major cause of death among HIV infected patients: role of hepatitis C and B viruses and alcohol. J Hepatol. 2005;42:799–805.
6. Reisler K. High hepatotoxicity rate seen among HAART patients. AIDS Alert. 2001;16:118–119.
7. Sulkowski MS, Thomas DL, Chaisson RE, Moore RD. Hepatotoxicity associated with antiretroviral therapy in adults infected with human immunodeficiency virus and the role of hepatitis C or B virus infection. JAMA. 2000;283:74–80.
8. Salmon-Ceron D, Rosenthal E, Lewden C, et al.. Emerging role of hepatocellular carcinoma among liver-related causes of deaths in HIV-infected patients: the French national Mortalite 2005 study. J Hepatol. 2009;50:736–745.
9. Joshi D, O'Grady J, Dieterich D, et al.. Increasing burden of liver disease in patients with HIV infection. Lancet. 2011;377:1198–1209.
10. Liaw YF, Sung JJ, Chow WC, et al.. Lamivudine for patients with chronic hepatitis B and advanced liver disease. N Engl J Med. 2004;351:1521–1531.
11. Sulkowski M. Hepatocellular carcinoma in HIV-infected patients comes of age: the convergence of epidemiology and treatment effectiveness. J Hepatol. 2009;50:655–658.
12. Berretta M, Garlassi E, Cacopardo B, et al.. Hepatocellular carcinoma in HIV-infected patients: check early, treat hard. Oncologist. 2011;16:1258–1269.
13. Brau N, Fox RK, Xiao P, et al.. Presentation and outcome of hepatocellular carcinoma in HIV-infected patients: a U.S.-Canadian multicenter study. J Hepatol. 2007;47:527–537.
14. Puoti M, Bruno R, Soriano V, et al.. Hepatocellular carcinoma in HIV-infected patients: epidemiological features, clinical presentation and outcome. AIDS. 2004;18:2285–2293.
15. Loko MA, Salmon D, Carrieri P, et al.. The French national prospective cohort of patients co-infected with HIV and HCV (ANRS CO13 HEPAVIH): early findings, 2006-2010. BMC Infect Dis. 2010;10:303.
17. Bruix J, Sherman M, Llovet JM, et al.. Clinical management of hepatocellular carcinoma. Conclusions of the Barcelona-2000 EASL conference. European Association for the Study of the Liver. J Hepatol. 2001;35:421–430.
18. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology. 2005;42:1208–1236.
19. Llovet JM, Bru C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis. 1999;19:329–338.
20. From the Centers for Disease Control and Prevention. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. JAMA. 1993;269:729–730.
21. Ozenne V, Bouattour M, Goutte N, et al.. Prospective evaluation of the management of hepatocellular carcinoma in the elderly. Dig Liver Dis. 2011;43:1001–1005.
22. Bruix J, Sherman M. Management of hepatocellular carcinoma: an update. Hepatology. 2011;53:1020–1022.
23. Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. 2005;353:487–497.
24. Clifford GM, Rickenbach M, Polesel J, et al.. Influence of HIV-related immunodeficiency on the risk of hepatocellular carcinoma. AIDS. 2008;22:2135–2141.
25. Vibert E, Duclos-Vallee JC, Ghigna MR, et al.. Liver transplantation for hepatocellular carcinoma: the impact of human immunodeficiency virus infection. Hepatology. 2011;53:475–482.
26. Kim JE, Kim YS, Rhim H, et al.. Outcomes of patients with hepatocellular carcinoma referred for percutaneous radiofrequency ablation at a tertiary center: analysis focused on the feasibility with the use of ultrasonography guidance. Eur J Radiol. 2011;79:e80–e84.
27. Patwardhan V, Paul S, Corey KE, et al.. Hepatocellular carcinoma screening rates vary by etiology of cirrhosis and involvement of gastrointestinal sub-specialists. Dig Dis Sci. 2011;56:3316–3322.
28. Tseng PL, Wang JH, Tung HD, et al.. Optimal treatment increased survival of hepatocellular carcinoma patients detected with community-based screening. J Gastroenterol Hepatol. 2010;25:1426–1434.
29. Subramanian A, Sulkowski M, Barin B, et al.. MELD score is an important predictor of pretransplantation mortality in HIV-infected liver transplant candidates. Gastroenterology. 2010;138:159–164.