|
Incidence of hepatocellular carcinoma in HIV/HBV-coinfected patients on tenofovir therapy: Relevance for screening strategies
|
|
|
After the initiation of TDF, the incidence of HCC remained stable over time, suggesting that an assessment of HCC risk at TDF start would be adequate to inform long-term individual HCC screening strategies.
Conclusions
As most HIV/HBV-coinfected individuals currently in care in high-income countries are non-cirrhotic with a suppressed HBV viral load on a TDF-containing regimen, it is of major importance to have reliable HCC incidence estimates to guide HCC surveillance. Our data suggest that the incidence of HCC is low in this group of patients, especially if TDF is initiated early during the course of disease. However, HCC events still occur in these patients and it will be important to further assess risk factors and derive predictive scores for HCC, tailored to HIV/HBV-coinfected populations.
Results
A total of 3,625 HIV/HBV-coinfected patients were included, of whom 72% had started TDF-containing ART. Over 32,673 patient-years (py), 60 individuals (1.7%) developed an HCC. The incidence of HCC remained stable over time among individuals on TDF, whereas it increased steadily among those not on TDF. Among individuals on TDF, the incidence of HCC was 5.9 per 1,000 py (95% CI 3.60-9.10) in cirrhotics and 1.17 per 1,000 py (0.56-2.14) among non-cirrhotics. Age at initiation of TDF (adjusted incidence rate ratio per 10-year increase: 2.2, 95% CI 1.6-3.0) and the presence of liver cirrhosis (4.5, 2.3-8.9) were predictors of HCC. Among non-cirrhotic individuals, the incidence of HCC was only above the commonly used screening threshold of 2 cases per 1,000 py in patients aged >45 years old at TDF initiation.
Highlights
•Over 32,673 patient-years of follow-up, 60 (1.7%) HIV/HBV-coinfected individuals developed HCC.
•In cirrhotic patients on TDF-containing antiretroviral therapy, the incidence of HCC was 5.90 per 1,000 patient-years.
•In non-cirrhotic patients on TDF-containing antiretroviral therapy, the incidence was 1.17 per 1,000 patient-years.
•HCC incidence stayed below the recognized screening threshold in non-cirrhotics who started TDF when <46 years old.
At initiation of TDF-containing ART, only age (aIRR per 10-year increase: 2.2, 95% CI 1.6-3.0) and the presence of liver cirrhosis (aIRR: 4.5, 95% CI 2.3-8.9) were significant predictors of the occurrence of HCC (Fig. 3, Table S1). Sex (aIRR: 0.7, 95% CI 0.2-2.4, ref: male) and ethnicity (aIRR: 1.6, 95% CI 0.6-4.6, ref: Caucasian) did not predict HCC incidence
|
|
|
|
|
|
|