Home Search Abstracts View Session E-mail Abstract Author


Session 132-Poster Abstracts
Hepatitis: Cirrhosis, Cancer, and Transplant
Thursday, 2-4 pm; Poster Hall
Paper # 685    
Screening for Hepatocellular Carcinoma in HIV/HCV Co-infected Patients: Impact on Staging, Therapy, and Survival
Marina Nunez*1, L Kikuchi2, P Barreiro3, M Nelson4, M E Vispo3, E Page4, R Fox5, E Bini6, M Sherman7, N Bräu8, and Atlantic Liver Cancer in HIV Study Group
1Wake Forest Univ Sch of Med, Winston-Salem, NC, US; 2Univ of Sao Paulo, Brazil; 3Hosp Carlos III, Madrid, Spain; 4Chelsea and Westminster Hosp, London, UK; 5Univ of California, San Francisco, US; 6New York Univ Sch of Med, NY, US; 7Univ of Toronto, Canada; and 8Mt Sinai Sch of Med, New York, NY, US

Background:  Current recommendations for Hepatocellular Carcinoma (HCC) screening in HCV-infected patients with cirrhosis (liver sonography every 6 to 12 months) are based on expert opinions, since there are no randomized controlled studies that show a survival benefit. No data are available on HCC screening in HIV/HCV-co-infected patients. This international cohort study of HIV/HCV co-infected patients with HCC compared outcomes in patients with and without prior HCC screening.

Methods:  HIV-infected patients with HCV-related HCC were retrospectively identified between 1992 to 2009 in 20 centers in North and South America and Europe. Each HCC diagnosis was confirmed using the 2005 AASLD criteria. Patients were considered screened if they initially presented with an abnormal AFP level or liver imaging study, and not screened if they presented with symptoms.

Results:  We identified 70 HIV/HCV co-infected patients with HCC, of whom 39 (55.7%) were screened. There was no significant difference between screened and non-screened patients in mean age (50.2 vs 53.3 years, =0.10), race/ethnicity (black, 44% vs 55%; white, 44% vs 26%; Latino, 13% vs 19%; =0.29), median HIV RNA level (<400 vs 895 copies/mL, =0.27), and median CD4+ cell count (308 vs 227 per mm3, =0.25). However, screened patients were more likely than non-screened patients to present with earlier Barcelona Clínic Liver Cancer (BCLC) stages A or B (69.2% vs 19.4%, <0.001), had a lower mean Cancer of the Liver Italian Program (CLIP) staging score (1.6 vs 2.5, =0.009), a lower mean Child-Turcotte-Pugh (CTP) score (6.6 vs 7.5, =0.027), had more commonly CTP stage A (61.5% vs 35.5%, =0.030), and they tended to more commonly receive effective HCC therapy (64.1% vs 41.9%, =0.064). After adjustment for lead time (calculated as 176 days using the tumor doubling time method), screened vs non-screened patients had a longer median survival (6.9 vs 3.8 months, =0.026).

Conclusions:  This retrospective study shows that a large proportion of HIV/HCV co-infected patients with HCC were not screened for the malignancy. Yet, screening for HCC among HIV/HCV-co-infected patients was associated with significantly earlier HCC stage, increased use of effective HCC therapy, and with improved survival after adjustment for lead time bias. These findings support the AASLD 2005 practice guidelines for HCC screening that can also be applied to HIV/HCV co-infected patients.