Mark A. Socinski, MD, discusses the nuances of the PACIFIC trial and ongoing research with durvalumab that could extend the reach of immunotherapy in early-stage lung cancer.
Durvalumab (Imfinzi) continues to display a significant overall survival (OS) and progression-free survival (PFS) advantage for patients with unresectable stage III non–small cell lung cancer (NSCLC), irrespective of PD-L1 expression, which should not be used to determine a patient’s eligibility for the agent, according to Mark A. Socinski, MD.
In 4-year findings from the PACIFIC trial, the OS rate was 49.6% with durvalumab in patients who did not have progressive disease after chemoradiation versus 36.3% with placebo. The PFS rates were 35.3% and 19.5%, respectively.1
“When you look at the approval of durvalumab after chemoradiation, the benefit is there in the [PD-L1–]high and [PD-L1–]low [populations]. The FDA [approval] in the United States is agnostic to PD-L1 status,” said Socinski.
Although durvalumab is also approved for use in Europe, the label is restricted to patients with at least 1% PD-L1 expression based on a small post hoc analysis that failed to show an OS benefit in patients with PD-L1 expression of less than 1% (HR, 1.14; 95% CI, 0.71-1.84).2
“Many of us think that that’s not the right thing to do, because you could not scientifically conclude that from a retrospective unplanned analysis where there was a lot of missing data,” said Socinski.
As such, ongoing research evaluating whether the addition of the PD-1 inhibitor to chemoradiation (NCT03519971) or stereotactic body radiation therapy (NCT03833154) can further improve outcomes for patients with stage III and I/II disease, respectively, will not use PD-L1 expression as an inclusion criterion.
In an interview with OncLive® during the 18th Annual Winter Lung Cancer Conference, a program hosted by the Physicians Education Resource® (PER®), LCC, Socinski, executive medical director of AdventHealth Cancer Institute, discussed the nuances of the PACIFIC trial and ongoing research with durvalumab that could extend the reach of immunotherapy in early-stage disease.
Socinski: PD-L1 is a pretty good biomarker to use. It’s not a perfect biomarker. If you look at large tumor biopsies or excisions, there will be parts of the tumor that are very PD-L1 positive, and then there’ll parts of the tumor that are PD-L1 negative. When you biopsy a tumor, just by sampling chance, you may biopsy the part of [the tumor] that doesn’t have any PD-L1, which we would call that PD-L1 negative even though that may not be truly reflective of the entire cancer.
It’s pretty clear from the phase 3 trials that have been done that PD-L1–negative patients still get very similar benefit [as PD-L1–positive patients] when you give immunotherapy with chemotherapy. Though, not so much when you give [immunotherapy] by itself. If you’re giving [immunotherapy] with chemotherapy, which in the PD-L1–negative population we would always use [immunotherapy] with chemotherapy, the PD-L1–negative patients do seem to benefit as much as the PD-L1–positive patients.
In stage IV disease, in the PD-L1–negative population, you would never use immunotherapy alone. You could always use [immunotherapy] with chemotherapy, and when you use [immunotherapy] with chemotherapy, the benefit in the PD-L1–negative population seems to be roughly similar to the benefit in the PD-L1–positive population.
One of the issues with interpreting the PD-L1 story from the PACIFIC trial is that at least one-third or so of the patients didn’t have tissue to measure PD-L1, so there are a lot of missing data there. Less than 60% of patients or so had PD-L1 [testing], so you don’t know the whole story.
PD-L1 [testing] was not a requirement for the study. That’s why many patients didn’t have it available from their tissue. In the trial, the investigators did look at PD-L1[–high patients], which AstraZeneca defined as greater than 25% versus the [PD-L1–low patients, defined as] less than 25%. Like we’ve seen with most [single-agent] immunotherapies, the higher you stain for PD-L1, the greater the benefit you get from immunotherapy.
The [United States] indication doesn’t say anything about [PD-L1]. However, the Europeans wanted data from the PACIFIC trial to a greater degree of granularity, and they specifically wanted to look at these less than 25% [PD-L1 expressors]. The issue that many of us have with that—and this actually came up in an advisory board that I was on earlier today—is that this was not a planned analysis. Up to 40% of the patients didn’t have PD-L1 tested. When they looked at this in a very small subset of patients who were PD-L1 negative, the benefit did not seem to be there for [durvalumab]. Therefore, the European Medicines Agency [EMA] labeled durvalumab only for PD-L1–positive patients.
Now, it certainly creates a hypothesis that maybe PD-L1–negative patients with stage III disease [don’t derive benefit from] single-agent immunotherapy, but I don’t think you can conclude that based upon that specific analysis. In Europe, you have to be PD-L1 positive to get consolidation durvalumab after chemoradiation, whereas in the United States, everyone gets it.
We recently had the 4-year update of the PACIFIC trial, and there continues to be separation and plateauing of the curves in favor of consolidation durvalumab. It really has been an important development in increasing cure rates. With standard chemoradiotherapy, the average person would typically live 2 to 2.5 years. Now, with the addition of immunotherapy, the average patient lives more than 4 years.
Therefore, there is this significant benefit that seems to last at least 3 to 4 years. It has been interesting over the past couple of years to see the updates of the PACIFIC trial at various major oncology meetings. One of the concerns early on was [whether] this early benefit [would] eventually go away. According to the latest update, which is out to 4 years, [the benefit is] not going away.
I’m not holding my breath. The one observation that was made with the PACIFIC trial was that the trial was initially designed such that you had to start immunotherapy within 2 weeks of finishing chemoradiotherapy. Chemoradiotherapy can be pretty tough on patients, and it was really slowed down by accrual. Patients just hadn’t recovered fast enough to start [durvalumab] within 2 weeks [of finishing chemoradiation]. Therefore, the investigators loosened up the starting time to out to 6 weeks. If you look at those patients that started [durvalumab] within 2 weeks, they actually did significantly better than the patients that started after 2 weeks. Now, the question is, were they just [fitter] patients who had lesser toxicity? Maybe they had smaller tumors, and so they had a smaller volume of radiotherapy that they needed. Or is there something magic about starting immunotherapy sooner rather than later?
Now, I’m not a good enough immunologist to make the argument that sooner is better, but people who are good immunologists can create a story to suggest that there is a biologic basis for starting [immunotherapy] earlier. And if starting [immunotherapy] earlier is better, which that subset analysis suggested, then maybe earlier [use of immunotherapy] during radiation is better.
I don’t know specifically about durvalumab. Studies are ongoing, but I think if you look at the big four [PD-1/PD-L1 inhibitors]: durvalumab, atezolizumab [Tecentriq], pembrolizumab [Keytruda], and nivolumab [Opdivo], they’re all in studies as neoadjuvant or adjuvant therapy. Some of those data should be coming out over the next year or so. I’m anticipating that we’re going to see benefits in the stage I and II surgical patients.
In limited-stage SCLC, we’ve always had or embraced the role of prophylactic cranial irradiation [PCI]. A number of studies addressed this, years ago, and then, what we refer to as the famous meta-analysis [looked] at the role of PCI in reducing the incidence or likelihood that the [disease] recurs in the brain, which was an issue. That translated into a survival benefit, not necessarily an individual studies, but in the meta-analysis.
Over the years, there has been controversy about the attendant toxicity. At face value, it doesn’t seem like a good idea to give radiation to your brain. There is a new technique called hippocampal sparing, which reduces some of the side effects. People are trying to think differently, because not every patient relapses in the brain. Therefore, are there strategies that one could use in place of PCI? One would be surveillance. If you did MRI scans at certain intervals, would that identify patients early on? If you identify [CNS disease] early on, then [could we] use [that] treatment in those patients to control the brain disease?
There are some nuances of managing CNS disease in the extensive-stage population too. Although there has been controversy about PCI in that population, it’s not something that we typically do on any frequency nowadays.
1. Faivre-Finn C, Vicente D, Kurata T, et al. Durvalumab after chemoradiotherapy in stage III NSCLC: 4-year survival update from the phase III PACIFIC trial. Ann Oncol. 2020;31(suppl 4):S1178-S1179. doi:10.1016/j.annonc.2020.08.2281
2. Paz-Ares L, Spira A, Raben D, et al. Outcomes with durvalumab by tumour PD-L1 expression in unresectable, stage III non-small cell lung cancer in the PACIFIC trial. Ann Oncol. 2020;31(6):P798-806. doi:10.1016/j.annonc.2020.03.287