GENEVA – The influenza vaccine may interact with immune checkpoint inhibitors in patients with lung cancer, results of a small study suggest.
Among 23 patients with non–small cell lung cancer (NSCLC) treated with a drug targeted against programmed death-1 (PD-1), the seasonal flu vaccine appeared to produce good serologic protection against infection, but at the possible cost of an increase in the rate of immune-related adverse events (IrAE), reported Sacha Rothschild, MD, PhD, of University Hospital Basel (Switzerland) at the European Lung Cancer Conference.
“Over 50% of patients overall had an immune-related adverse event, and that’s certainly higher than what we have seen in all the studies, and it’s also clearly higher than what we see in our daily clinical practice, especially with grade 3/4 toxicity,” he said in an interview at the meeting.Among 23 patients with lung cancer treated with a PD-1 inhibitor, 12 (52.2%) had one or more IrAEs. In contrast, the most frequent IrAE in a key registration trial for nivolumab (Opdivo) was skin rash, which occurred in 9% of patients (N Engl J Med. 2015 Jul 9;373:1627-39).
“It’s a very small study, but it raises some concern that there might be an interaction between the vaccine and PD-1 blockade,” Dr. Rothschild said.
To see whether blocking the PD-1/PD–ligand-1 (PD-L1) axis might induce an overactive immune response, the investigators prospectively studied 23 patients with NSCLC who were undergoing treatment with a PD-1 inhibitor – 22 with nivolumab and 1 with pembrolizumab (Keytruda) – who were also vaccinated with a trivalent influenza vaccine in October or November 2015. They used the partners of the patients, also vaccinated, for an age-matched cohort of healthy controls.
The investigators looked at antibody titers against flu strains covered by the vaccine, measured inflammatory chemokines and assessed the vaccine’s safety and the frequency of IrAEs.
None of the patients came down with the flu during the 2015-2016 season. There were no major differences over time in the generation of antibodies against all three viral strains tested.
However, at both 30 and 60 days after vaccination, a hemagglutination inhibition assay showed slightly elevated antibody titers among patients, compared with controls. Antibody titers against H1N1 virus also appeared to increase somewhat more rapidly among patients than among controls, the authors found.
The patients appeared to tolerate the vaccine well, and no serious adverse events were reported within 30 days of vaccination.
When they looked at the incidence of IrAEs, however, the investigators found that six patients had grade 1 or 2 IrAEs, and six had grade 3 or 4 events.
The events included skin rash and arthritis in three patients each, colitis and encephalitis in two patients each, and hypothyroidism, pneumonitis, and neuropathy in one patient each.
“We looked into inflammatory chemokines to understand if there was a high rate of systemic inflammation, and we didn’t find any differences in this regard. So far, we have no clue about why the immune-related adverse event rate in this group is higher,” Dr. Rothschild said.
Although the sample size was small, the IrAE effect they saw was large enough to warrant concern, and it should be studied in a larger population sample, he said.
Egbert Smit, MD, PhD, of the Netherlands Cancer Institute in Amsterdam, who was not involved in the study, commented that “it shows how much we still have to learn about the optimal use of checkpoint inhibitors in lung cancer patients. The study is important as it is the first to investigate the impact of influenza vaccination in such patients, and there is a hint that we actually put them at increased risk for serious toxicities, including encephalitis. However, until we have data on a larger cohort, preferably in a controlled, prospective study, in my institution, we advocate influenza vaccination irrespective of concurrent treatment with immune-checkpoint inhibitors.”
The study was supported by institutional funding. The investigators and Dr. Smit reported no relevant conflicts of interest.