Pembrolizumab monotherapy vs. EXTREME
For the coprimary endpoint of OS in the CPS 20 or greater population, pembrolizumab was associated with significantly better survival than EXTREME at both the 12- and 24-month time points (56.9% vs. 44.9%, and 38.3% vs. 22,1%, respectively). After a minimum follow-up of 17 months, the median OS was 14.9 months with pembrolizumab, vs. 10.7 months for EXTREME. The hazard ratio (HR) for death with pembrolizumab was 0.61 (P = .0007).
The median OS in the CPS 1 or greater population was 12.3 months and 10.3 months, respectively (HR 0.78, P = .0086).
There were no differences between the arms in PFS, however, either in the CPS 20 or greater or CPS 1 or greater populations.
Although, as noted, response rates did not differ between the groups, the median duration of response was 20.9 months with pembrolizumab in both the CPS 20 and CPS 1 populations, compared with 4.2 and 4.5 months, respectively, for EXTREME.
Treatment-related adverse events of any grade occurred in 58% of patients in the monotherapy arm, vs. 96.9% in the EXTREME arm. Fatal adverse events occurred in 1% vs. 2.8%, and events leading to drug discontinuation occurred in 4.7% vs. 19.9%, respectively. There were more immune-mediated events in the pembrolizumab arm, including one death (from pneumonitis) vs. no deaths from immune-related causes in the EXTREME arm.
Pembrolizumab plus chemo vs. EXTREME
The combination of pembrolizumab was also superior to EXTREME in the total population, with 12- and 24-month OS rates of 53% vs. 43.9%, and 29% vs. 18.7%, respectively. The median OS was 13 months with the pembrolizumab/chemo combination, vs. 10.7 months for EXTREME, translating into an HR of 0.77 (P = .0034). In this analysis as well as in the pembrolizumab monotherapy combination, there was no difference in PFS or response rates, but responses in the pembrolizumab arm were more durable.
In this comparison, treatment-related adverse events were generally similar between the groups, although there were 10 treatment-related deaths with pembrolizumab, compared with eight in the EXTREME arm.
There was one immune-related death, from pneumonitis, in the pembrolizumab arm, vs. none in the EXTREME arm. Hypothyroidism, pneumonitis, hyperthyroidism and colitis were more frequent with pembrolizumab, whereas infusion reactions and severe skin reactions were more frequent with EXTREME.
“There are further analyses of biomarker and clinical predictors that will be forthcoming from this study, and these may eventually optimally guide the choice of whether to administer pembrolizumab alone or in the novel combination, Dr. Burtness said at the briefing.
“What’s extremely important also is that for a subgroup of patients with high expression of PD-L1, we can probably remove the cisplatin and have a good outcome with immunotherapy alone,” Dr. Machiels said.
He said that more work needs to be done to determine which patients are most likely to benefit from immunotherapy in the recurrent/metastatic setting, and “we have to see how we can now bring this active drug to the curative treatment of the patient, in combination with chemoradiation.”
The study was funded by Merck Sharp & Dohme. Dr. Burtness disclosed being and advisory board member and receiving travel expenses from MSD and others. Dr. Machiels disclosed speaker honoraria, travel expenses, and an uncompensated advisory role with MSD.
SOURCE: Burtness B et al. ESMO 2018. Abstract LBA8_PR.