Original Title: Detección precoz de hepatocarcinoma en pacientes de riesgo

Authors' objectives: The objective of this report is to evaluate the effectiveness, cost-effectiveness and budgetary impact, as well as the ethical, social, legal and organizational implications of early detection and diagnosis programs of HCC by abdominal ultrasound in patients at risk (i. e., cirrhosis of any etiology, HBV and HCV).

Authors' results and conclusions: RESULTS: A total of 2402 references were found in the electronic databases, and 2059 were considered after the elimination of duplicates. A second Pubmed search was then carried out with a less specific strategy for the same time period, in which 492 references were obtained that were not duplicated with the first search. Of the total number of titles obtained, 173 references were selected for reading to full text, after which 60 studies were included. Seven new studies were identified by handsearching. Therefore, 67 studies were finally included: 2 RS, 5 RCTs and 60 observational studies. The studies that evaluated the diagnostic performance of ultrasound compared to AFP reported very high heterogeneity values, so no statistical contrasts were made between the values of sensitivity and specificity of the ultrasound versus AFP. The difference observed between both techniques is reflected in the sensitivity, which is considerably greater with ultrasound, while the specificity shows very similar values. Studies comparing ultrasound versus ultrasound and AFP showed that the addition of AFP did not increase the sensitivity of the ultrasound. On the other hand, the results of the comparison between ultrasound and CT also showed no significant differences in early detection. However, studies that compared ultrasound with MRI showed that MRI reflects lower sensitivity but greater specificity than ultrasound. In reference to effectiveness, the results of this study are clearly favorable to surveillance, in line with other studies that have also offered significant and intense effects in favor of surveillance. Fifteen economic evaluations were identified. In all studies, any surveillance strategy was more effective and costly than no surveillance. In patients with cirrhosis, the combination of AFP and ultrasound appears to be more expensive and more effective than just ultrasound; the higher frequency of testing, ultrasound every 6 months (with or without AFP) versus these same tests every 12 months, is also more expensive and more effective; higher cost tests such as MRI or CT are more effective than ultrasound but also more expensive if used for HCC screening. In studies in patients with cirrhosis, ICERs are heterogeneous given the variety of technologies compared; only three studies identified dominant strategies. Two studies evaluated surveillance in HBV patients in two specific populations, native Alaskan population and Thai population. The results show that frequent ultrasound/AFP surveillance is more expensive and more effective than individual and less frequent tests. In the base case, ICER is estimated at 9,385.94 €/LYG and 11,674.09 €/QALY in patients with cirrhosis due to HCV, 12,199 €/LYG and 17,043.47 €/QALY in patients with cirrhosis due to HBV, and 17,390.97 €/LYG and 24,239.96 €/QALY in patients with alcoholic cirrhosis. All these ratios are below the cost-effectiveness thresholds estimated for Spain. Probabilistic sensitivity analysis shows some uncertainty although most simulations show ultrasound every 6 months is cost-effective or even more effective and less costly than not monitoring. The cost of a surveillance program (ultrasound every 6 months) in patients with cirrhosis, and assuming 100% adherence, is estimated in 904,511,979 € in 5 years for the NHS. CONCLUSIONS: - The current evidence on the effectiveness of ultrasound surveillance programs, with or without AFP, is of low or very low quality. However, future RCTs are unlikely to be carried out, given the ethical problem of randomizing a risk group to non-surveillance. In observational studies, it is desirable to carry out prospective studies of good methodological quality, but it is difficult to find homogeneous comparison groups in their clinical characteristics, since generally the people screened will be those at greatest risk. - With the low level of quality mentioned above, the available results are favorable to surveillance by detecting a higher rate of early stage tumors, and resulting in a higher rate of application of curative treatments and survival at 3 and 5 years. - There is no evidence for the differential efficacy of screening programs in relation to the population at risk (cirrhosis, HBV or HCV). - There is moderate quality evidence that screening every 3 months in patients with cirrhosis is no more than 6 months. There is very low quality evidence that screening every 6 months is superior to annual screening. - There is no quality evidence that the addition of AFP improves diagnostic performance or health outcomes obtained by screening with ultrasound alone. - There is very low quality evidence that CT scan is not superior to ultrasound in diagnostic performance. - There is very low quality evidence that MRI shows better diagnostic performance than ultrasound. - Surveillance for the early detection program of HCC is more effective and more costly than not monitoring the patient at risk. - The combination or increased frequency of testing is more effective and more costly than monitoring by individual or less frequent testing. - The literature review offers heterogeneous results so that there is no indication that some strategies are more cost-effective than others consistently. - The economic evaluation conducted for this report found that a surveillance program for early detection of HCC consisting of ultrasound every 6 months is cost-effective compared to no surveillance in patients with cirrhosis of the liver, due to HCV, HBV or alcoholic cirrhosis, from the Spanish NHS perspective. - The cost of ‘no surveillance’ is lower than the cost of ultrasound every 6 months in the whole NHS.

Authors' methods: The following electronic databases were consulted: Medline (1946-March 2017), Embase (1974-March 2017), Cochrane Library (1995-March 2017), CRD (1994-march 2017), CINAHL (1982-March 2017), Pubmed (March 2012-March 2017), BVS (1982-March 2017), Medes (2001-March 2017), ICTRP (-March 2017). Two reviewers performed the entire selection process of studies and in case of doubt or disagreement between them a third reviewer was consulted, who revised the protocol criteria and tried to reach a consensus with the other two reviewers. We selected systematic reviews with and without meta-analysis , observational studies were selected that examined the efficacy of baseline techniques for early detection of HCC versus other radiological techniques, determination of alpha-fetoprotein (AFP) and other existing detection and early diagnosis procedures for HCC in patients at risk of HCC (cirrhosis, hepatitis (VHB). We also included systematic reviews covering ethical, organizational, social and patient aspects, opinions and editorials regarding the population, intervention and comparison mentioned above. The study selection was limited to those published in English and Spanish. Once the studies that met the inclusion criteria were identified, the contents of interest were fed into data extraction sheets. The instruments used to assess risk of bias in the case of Systematic review (SR) were the AMSTAR Scale, the QUADAS-2 scale for diagnostic performance studies, the Cochrane Collaboration tool for randomized controlled trials (RCTs) and non-randomized controlled trials (NRCTs), while for observational studies the Scottish Intercollegiate Guidelines Network (SIGN) criteria were applied. For the systematic review of economic evaluations we selected complete economic evaluations, developed alongside primary studies (clinical trials for example) or economic models. The studies should include life years gained (LYG) or quality-adjusted life years (QALYs) as an effectiveness measure, and report the costs of each alternative in comparison or the values of incremental cost-effectiveness ratios (ICERs). The economic evaluations were evaluated by means of the López-Bastida et al. (2010) questionnaire. A full economic evaluation was performed comparing LYG, QALYs and costs of HCC surveillance versus no surveillance in cirrhosis patients. The literature review and expert opinion helped define the clinical course and resource use of each alternative in comparison. Sensitivity and specificity values are derived from the meta-analysis carried out in this report. A Markov model was built for a life-time horizon of the patient's entire life. The analysis was performed for a hypothetical cohort of 100,000 people with cirrhosis and from the NHS perspective, therefore only direct health costs covered by the NHS in Spain were included. Costs were expressed in euros of 2017. The costs and effectiveness of the two alternatives and the ICER were estimated. Deterministic sensitivity analysis and probabilistic sensitivity analysis were performed. The results are presented separately for the three types of cirrhosis (HCV, HBV, and alcoholic cirrhosis). A budget impact analysis was conducted, estimating the cost for the NHS of three scenarios: no surveillance; surveillance with biannual ultrasound and 100% adherence; surveillance with biannual ultrasound and 77% adherence. Gross cost is estimated for a time horizon of 5 years. Based on population statistics and assumptions, the number of patients with cirrhosis is estimated for each cause and the costs are presented separately for HCV, HBV and alcoholic cirrhosis.

Project Status: Completed

Year Published: 2017

URL for published report: https://www3.gobiernodecanarias.org/sanidad/scs/content/3aa58f0e-2623-11e9-af40-97fc2ea86ebc/SESCS_2017_Hepatocarcinoma.pdf

Requestor: Ministry of Health

English language abstract: An English language summary is available

Publication Type: Full HTA

Country: Spain

Organisation Name: Canary Health Service

Contact Address: Dirección del Servicio. Servicio Canario de la Salud, Camino Candelaria 44, 1ª planta, 38109 El Rosario, Santa Cruz de Tenerife

Contact Name: sescs@sescs.es

Contact Email: sescs@sescs.es