BARCELONA, Spain — The long-term nucleoside and nucleotide analogue treatment of patients with chronic hepatitis B is not associated with an increased risk for extrahepatic cancers, according to results from a new study that led to some confusion.
The findings, presented here at the International Liver Congress 2016, were publicized in a press release that raised a false alarm by warning that long-term treatment is linked to colorectal and cervical cancer, even though the study came to the opposite conclusion.
A subsequent analysis of the data flipped the outcome, explained investigator Grace Wong, MD, from the Chinese University of Hong Kong.
"I agree that the press release appears to be an overinterpretation," acknowledged European Association for the Study of the Liver (EASL) Vice-Secretary Tom Hemming Karlsen, MD, PhD, from Oslo University Hospital.
How did the press release get it so wrong, warning journalists and the public of a cancer risk when there may be nothing to worry about? Candice Debleu, from Ruder Finn Public Relations Agency, explained that the announcement was written based on outdated abstract data.
"The press release data align to the abstract data," Debleu told Medscape Medical News. But "the abstracts were submitted to the congress in November 2015," she added.
Dr Wong acknowledged that the study finding did change after a review by her team's statisticians, who suggested they apply the Bonferroni correction to account for the multiple comparisons for 12 different malignancies in the same analysis.
They then set a new threshold for significance at P < .004 in place of P = .05 to help offset any falsely significant findings.
"I think it is important that the press releases reflect the most recent data being presented. We should consider asking the authors next year to share the most recent data that will be presented in the scientific sessions to ensure there are no discrepancies," Debleu explained.
The researchers retrospectively identified patients with chronic hepatitis B treated from 2000 to 2012 with one or more analogue — including lamivudine, adefovir, entecavir, telbivudine, and tenofovir — using data from the Hospital Authority.
Patients with pre-existing malignancies and coinfection with other viruses were excluded, but premalignant conditions, such as gastrointestinal polyps, were treated as disease modifiers, said Dr Wong.
Investigators used landmark analysis to control for potential "immortal time bias," and they used propensity-score weighting to control for baseline characteristics that might influence the probability of receiving nucleoside or nucleotide analogue therapy.
Overall, 44,494 patients with chronic hepatitis B were included in the analyses: 4782 who were treated with analogues and 39,712 who were untreated.
After a median follow-up of 4.4 years, incidence rates of cancer in the chronic hepatitis B population were either similar to or lower than rates in the general population, "except for hepatocellular carcinoma, which makes sense," Dr Wong said.
Rates of gastric, esophageal, lung, pleural, breast, retroperitoneal, urinary, renal, brain, and other cancers were not elevated in patients with hepatitis B. But there were two exceptions. Lymphoma was more frequent in patients with chronic hepatitis B than in the general population (8.5 vs 12.3 per 100,000 person-years), as was cervical cancer (18.0 vs 6.4 per 100,000 person-years).
Treatment was not associated with higher rates of cancer overall (hazard ratio [HR], 1.01; P = .899), or of individual cancers. The trends toward increased colorectal cancer (HR, 2.17; P = .029) and cervical cancer (HR, 7.33; P = .007) did not remain significant after adjustment.
A subgroup analysis showed that women had a higher risk for colorectal cancer than men (weighted HR, 5.78; P = .003), but Dr Wong emphasized that "the absolute numbers of colorectal cancer were small — just five cases in the treated group and 15 in the untreated group."
In fact, the overall risk for colorectal cancer was lower in the hepatitis B patients than in the general population (incidence rate, 28.8 vs 63.8 per 100,000 person-years), "Even among treated patients, the risk was not higher than in the general population," she pointed out. No additional colorectal and cervical cancer screening should be offered, beyond what is normally recommended, she concluded.
Nucleoside and nucleotide analogues "have been used for many years without any major problems," said Dr Wong. "Our study showed that the treatment does not increase the risk of cancers as a whole."
The first-line therapy for chronic hepatitis B is "very, very safe" in the long term, said Robert Gish, MD, medical director of the Hepatitis B Foundation in Doylestown, Pennsylvania.
"The benefits of preventing liver cancer, cirrhosis, and liver transplant far outweigh any theoretical risk of other cancers," Dr Gish added.
The revised numbers still raise some questions about risks for women, but "we believe it would be more objective to state that such an increase in risk requires independent confirmation," Dr Wong explained.
"This large-scale study adds nuances to the discussion about the long-term adverse effects of nucleoside or nucleotide analogue treatment, but further data are needed before conclusions on cancer risk can be drawn," said Dr Karlsen after learning of the revised results.
"The colon cancer risk in women was a statistical event at most, and not clinically significant," said Dr Gish. The risk for significant adverse effects with nucleoside or nucleotide analogues is "less than 1% per year."
Dr Wong reports being a member of the advisory committee for Otsuka and Gilead, and a speaker for Abbott, AbbVie, Bristol-Myers Squibb, Echosens, Furui, Gilead, Janssen, Otsuka, and Roche. Dr Karlsen, Dr Gish and Ms Debleu have disclosed no relevant financial relationships.
International Liver Congress (ILC) 2016: Abstract PS052. Presented April 15, 2016.
Medscape Medical News © 2016 WebMD, LLC
Send comments and news tips to firstname.lastname@example.org.
Cite this: Release Raises False Alarm Over Hep B Meds and Cancer Risk - Medscape - Apr 22, 2016.