PD-L1 Testing in Stage III NSCLC & General Education
Insights From: John G. Gosney, MD, FRCPath, Royal Liverpool and Broadgreen University Hospital NHS Trust
John R. Gosney, MD, FRCPath: PD-L1 [programmed death-ligand 1] testing, of course—that is, assessing the proportion of tumor cells expressing PD-L1 protein on their surface—is now absolutely standard in the profiling of non–small cell lung cancer. And sometimes it’s a complementary test in which it’s merely informative. Sometimes it is a companion test, which means it’s mandatory in order to allow a prescription of an immune modulator in a particular situation. Interestingly, that’s applied almost only to advanced-disease, stage IV disease. Now, of course, we have a license for the use of durvalumab in stage III disease, admittedly beyond radiochemotherapy when disease is stable. But this is a development because it has essentially brought the need for PD-L1 testing into an earlier tubal stage. And we didn’t have this before.
And it’s extremely important in view of the fact that the European Medicines Agency has of course been very key to see PD-L1 testing in stage III non–small cell lung cancer, so that a level of 1% can be sought as a cutoff to the use of durvalumab. And in any situation in which tissue is available, that test has to be done. So we now have to test every patient with stage III non–small cell lung cancer for PD-L1 expression.
If the practice has been to test all patients with non–small cell lung cancer, or at least those maybe beyond stage I at diagnosis by reflex testing, this is not a problem. That cohort of patients with stage III disease, which is about one-third of the total, is already tested for. The problem is when testing has been confined to stage IV advanced disease, and those patients require testing retrospectively. This is when the problems can arise when the tissue might have been archived or might have been moved to another hospital. And those problems need addressing. But essentially, yes, it’s extremely important—as far as possible—to have patients with stage III non–small cell lung cancer, to have that tumor assessed with PD-L1 expression.
It’s important to realize that unlike genomic drivers—EGFR mutation, ALK rearrangement—PD-L1 expression is a continuum. So in any given tumor, any biopsy of any given tumor, you might have no cells expressing PD-L1 or PD-L1 and anywhere in between. So essentially what you consider positive depends on where you decide to put the bar. This is essentially a statistical and scientific construct. And the common cutoffs are at 50% and at 1%. And in any given group of patients with non–small cell lung cancer, about two-thirds will have expression levels greater than 1% and about a third greater than 50%. So essentially three-thirds.
And in terms of prescribing durvalumab, for example, in patients with stage III disease, the cutoff that has been mandated by the European Medicines Agency is at 1%, not 50%. And that is the current situation. But those cutoffs are movable and they change, and they are dependent upon the particular immune modulator in question and the context—in terms, for example, of whether you’re using or wishing to use the immune modulator first line or second line.
Communication, and indeed education—not only within a team of health care professionals who are managing patients with non–small cell lung cancer but beyond that—is hugely important. When you have a multidisciplinary team meeting, a tumor board meeting, then clearly there’s dialogue. And the people within that group are constantly aware of the problems of the pitfalls, of the difficulties, and so on. But of course you don’t have in the room on those occasions often necessarily the clinicians who might have obtained the biopsy in the first place, maybe at a peripheral hospital, nor do you have the pathologist who’s been responsible for the initial diagnosis and work-up.
And it’s very important that those health care professionals—beyond the immediate team—are engaged, educated, and involved. And this is not always easy to do. And you often find that within a group of highly specialized individuals, when knowledge is very, very high and people are up to date and engaged, you can quickly forget that that is not the norm with the major sensors. And that communication—and bringing in these different personnel, and making sure they feel part of what is a network that needs to function in order to benefit the patient most—is working. That really has to be worked out.
For example, it’s very important for me as a pathologist to constantly remind—in a diplomatic way—those clinicians who are obtaining the material for me, which is largely physicians and radiologists, how they are of crucial importance. And they might not always be in the room at the time. I might not see them. But that sort of communication is also important, as is this communication with pathologists in peripheral hospitals who need to be encouraged not to waste tissue, not to do unnecessary immunochemistry, and so on. So that broader education is really of very great importance. They have to feel positive too, as well as the people in the tumor board and the specialist center.