In a case-control study of patients with cirrhosis, screening for hepatocellular carcinoma up to 4 years prior to diagnosis was not associated with lower mortality.
Similar proportions of cases and controls underwent screening with abdominal ultrasonography, serum alpha-fetoprotein (AFP) testing, or both, reported Andrew M. Moon, MD, MPH, of the University of North Carolina at Chapel Hill, and his associates. “There was also no difference in receipt of these screening tests within 1, 2, or 3 years prior to the index date,” they wrote. The report was published in. The findings “[suggest] that either these screening tests or the currently available treatments [for liver cancer], or both, are suboptimal and need to be improved.”
Because cirrhosis significantly increases the risk of hepatocellular carcinoma, the American Association for the Study of Liver Diseases, the European Association for the Study of the Liver, and the Asian Pacific Association for the Study of the Liver recommend screening cirrhotic patients every 6 months with abdominal ultrasonography with or without concomitant serum AFP. But nonliver societies have not endorsed this approach, citing a lack of high-quality data. One problem is that studies have compared patients whose liver cancer was diagnosed by screening with those diagnosed after they became symptomatic, which creates a lead-time bias that inherently favors screening, Dr. Moon and his associates noted.
To help fill the evidence gap, they identified 238 patients from the Veterans Affairs health care system who had died of hepatocellular carcinoma between 2013 and 2015 and who had been diagnosed with cirrhosis at least 4 years beforehand. They compared these cases with an equal number of patients with cirrhosis who had been in VA care for a similar amount of time and had not died of hepatocellular carcinoma. Cases and controls were matched by etiology of cirrhosis, year that cirrhosis was diagnosed, race, age, sex, Model for End-Stage Liver Disease score, and VA medical center. The researchers identified screening tests by reviewing blinded medical charts.
There were no significant differences in the proportions of cases and controls who underwent screening ultrasonography (52.9% versus 54.2%, respectively), screening serum AFP (74.8% versus 73.5%), either test (81.1% versus 79.4%), or both tests (46.6% versus 48.3%) within 4 years of the index date or the matched control. The result was similar after potential confounders were controlled for and when examining shorter time frames of 1, 2, and 3 years.
It was unlikely that these results reflect delayed diagnosis of liver cancer or a lack of treatment within the VA system, the experts wrote. A total of 51.3% of cases were diagnosed with Milan criteria, which exceeds the proportion in the national Surveillance, Epidemiology, and End Results registry, they noted. None of the fatal cases underwent liver transplantation, but 66.8% received other treatments for liver cancer.
Funders included the National Institutes of Health and the Veterans Affairs Clinical Science Research & Development. The investigators reported having no conflicts of interest.
SOURCE: Moon AM et al. Gastroenterology. 2018 Jul 5. doi: .