BARCELONA — The advent of immunotherapy for advanced non-small cell lung cancer (NSCLC) has transformed an almost universally fatal disease into something akin to a chronic condition for many patients, long-term data from randomized trials suggested.
Pooled data from two randomized trials showed a 5-year overall survival (OS) of 13% for patients treated with nivolumab (Opdivo), as compared with 2% for patients treated with docetaxel. The 3-year results from a trial comparing single-agent pembrolizumab (Keytruda) and chemotherapy showed almost a 20-point absolute difference in favor of the patients treated with the PD-1 inhibitor.
In the nivolumab study, patients who survived without disease progression at 2 or 3 years had a good chance of surviving to 5 years, said Scott Gettinger, MD, of Yale Cancer Center in New Haven, Connecticut, at the World Conference on Lung Cancer (WCLC).
“Patients without disease progression at 2 and 3 years after starting nivolumab had a 60% and 78% chance of remaining without progression at 5 years,” he said during a WCLC press briefing. “There were no baseline clinical or tumor characteristics that clearly distinguished long-term survivors receiving nivolumab.”
“Of note, at 5 years, 10% of nivolumab survivors had been off [the] study drug for variable degrees of time, had not progressed, and had not received subsequent therapy,” he added. “We clearly see benefit long after they finish a course of immunotherapy, or have at least stopped for some reason. With 5 years minimum follow-up, no new safety signals were identified for nivolumab.”
Despite the favorable long-term results, lung cancer specialists avoided using the term “cure.”
“This is something we never thought to be possible 5 years ago in non-small cell lung cancer,” said Martin Reck, MD, of the Lung Clinic Grosshansdorf in Germany. “The question is whether we can really talk about cure of lung cancer after 5 years. This has been our traditional view, if we had progression-free survival for 5 years or longer we may assume the patient is cured. However, we have data that when we biopsy these patients, we still see viable tumor in some of them.”
“I would just say we are controlling the tumor, but we have to think about how we manage active treatments. I think we cannot manage these patients without any active treatment or surveillance. I think we really have to redefine what we mean by ‘cure’ in lung cancer,” he said.
On the other hand, Gettinger said autopsies of some long-term survivors showed no evidence of viable tumor.
Gettinger reported 5-year follow-up data from the CheckMate 017 and 057 studies comparing nivolumab and docetaxel versus docetaxel alone. The pooled analysis comprised data for 854 randomized patients. The results showed large differences in favor of the nivolumab arm beginning with the first landmark survival analysis:
- 1-year OS: 48.0% vs 34.3%
- 2 years: 26.9% vs 13.5%
- 3 years: 17.1% vs 8.1%
- 4 years: 14.2% vs 4.6%
- 5 years: 13.4% vs 2.6%
Median OS also favored the nivolumab arm, 11.1 versus 8.1 months, representing a 32% reduction in the survival hazard (95% CI 0.59-0.78).
Reck presented the 3-year follow-up results from the KEYNOTE-024 trial comparing pembrolizumab and platinum-based chemotherapy in 305 patients with metastatic NSCLC. Similar to the nivolumab data, the advantage of immunotherapy over chemotherapy emerged within the first year. After a median follow-up of 11.2 months, patients treated with pembrolizumab already had a 50% lower risk of disease progression or death and a 40% lower risk of death within the first year.
At 2 years, 51.7% of pembrolizumab-treated patients remained alive as compared with 34.2% in the chemotherapy group, and 3-year survival was 43.7% versus 24.9%. After a median follow-up of 44.4 months and a minimum of 3 years, the median OS was 26.3 months with pembrolizumab versus 14.2 months with chemotherapy.
The survival benefit with pembrolizumab occurred with a reduced risk of treatment-related adverse events (TRAEs, 77% vs 90%, all grades), grade 3-5 TRAEs (31% vs 53%). Serious TRAEs and rates of events leading to discontinuation or death occurred in a similar proportion of patients. As expected, immune-mediated adverse events and infusion reactions occurred more often with pembrolizumab (34%) than with chemotherapy (5%).
The CheckMate 017 and 057 studies were supported by Bristol-Myers Squibb (BMS).
Gettinger disclosed relevant relationships with BMS and Nektar.
The KEYNOTE-024 trial was supported by Merck.
Reck disclosed relevant relationships with Roche, BMS, AstraZeneca, Merck Sharp & Dohme, Merck, Boehringer-Ingelheim, Celgene, Lilly, Novartis, Abbvie, and Pfizer.