Page 236 - ebook HCC

GENEVA, SWITZERLANDA, SWITZERLAND
PROGRAMME AND ABSTRACTSAND ABSTRACTS
EASL HCC SUMMITHCC SUMMIT
234
235
234 PROGRAMME GENEV EASL 235
FEBRUAR
FEBRUARY 13 - 16, 2014Y 13 - 16, 2014
Poster Board Number C25
ALPHA-FETOPROTEIN CHANGES AS
SURVEILLANCE TEST FOR HEPATOCELLULAR
CARCINOMA: RESULTS OF A CASE-CONTROL
STUDY

Fabio Conti , Maurizio Biselli , Marta Frigerio , Annagiulia Gramenzi , Results: In TG AFP increased from T-12 to T0 in HCC (p=0.001) whereas remained
1
1
1
1
Marco Dall’Agata , Maura D’Angelo , Edoardo G. Giannini , Fabio Farinati , stable in controls (p=0.126). AFP was also significantly higher in HCC than controls
1
3
2
1
Paolo Del Poggio , Mauro Bernardi , Franco Trevisani , Pietro Andreone 1 at each time point (p<0.001). At multivariate analysis AFP T0 and AFP Δ6+ were
1
1
4
1 Department of Medical and Surgical Sciences, University of Bologna, Bologna, independently associated with the diagnosis of HCC (Odds ratio: 1.031 and 2.402,
2 Department of Internal Medicine, University of Genoa, Genoa, Department of Surgical respectively). The area under the curve of AFP T0 was 0.76 and its best cut-off was 10
3
4
and Gastroenterological Sciences, University of Padua, Padua, Division of Medicine, ng/ml (AFP10) with a sensitivity of 66.3% and specificity of 80.6%. In order to improve
Treviglio-Caravaggio Hospital, Treviglio, Italy the performance of the surveillance test we combined AFP10 with AFP Δ6+ obtaining a
sensitivity of 80% (95%CI 74.3-84.8%), with NPV of 86.2% (95%CI 81-90.2%). Appling
Corresponding author’s e-mail: conti.fabio1@libero.it this test at the 3% HCC prevalence reported in literature NPV raised to 99% (95%CI
96.5-99.8%). VG confirmed 10 ng/ml as AFP T0 best cut-off value with similar sensibility
(66.7%) and an increased specificity (88.9%) versus TG. The combination of AFP 10 and
Introduction: In patients with cirrhosis surveillance for hepatocellular carcinoma (HCC) AFP Δ6+ allowed to obtain a sensitivity up to 80.6% (95%CI 73-86.5%) with an high NPV
is highly recommended. value (90.5%, 95%CI 84.2-94.6).
Aims: In this case-control study we evaluated the efficiency of overtime elevation of alpha- Conclusions: Combination of AFP as a single dosage with a cut-off of 10 ng/mL and the
fetoprotein (AFP) as surveillance test for HCC. presence of a positive difference between the values of AFP in the last 6 months increased
accuracy of HCC identification in cirrhotic patients.
Methodology: we recruited 80 patients newly diagnosed with HCC (cases) from
January 2000 to February 2009 during a semiannual surveillance program based on
ultrasonography and AFP measurement at 4 referral centers. They were matched 1:2
CLINICAL POSTER ABSTRACTS group, TG). Validation group (VG) consisted of 36 patients newly diagnosed with HCC CLINICAL POSTER ABSTRACTS
for age, gender, etiology and Child-Pugh class (at time of case diagnosis) with cancer-
free controls undergoing the same surveillance program at the same centers (training
at only 1 center matched 1:3 with cancer-free controls from March 2009 to May 2013.
AFP values at 12 (T-12) and 6 (T-6) months before and at the time of HCC diagnosis
(T0) and a positive difference between AFP at T0 and T-6 (Δ6+) or T-12 (Δ12+) were
collected. To align AFP measurements we considered as T0 for controls the date of
AFP measurement closest to the date of HCC diagnosis of the corresponding case.
Regression analysis was used to assess the association between AFP and HCC.