Strategic Approaches to Biomarker Testing in NSCLC: Optimal Timing

Transcript:

Charu Aggarwal, MD, MPH: I’ll move to Dr [Martin] Dietrich. Understanding that biomarker testing is important, and knowing all the advances that are occurring in our field, walk us through when this should be performed for patients with lung cancer, and how do you incorporate it into your management paradigm?

Martin Dietrich, MD, PhD: I think that has recently changed. We used to test everybody with locally advanced and metastatic disease, and now we have perioperative therapies swooping in, improving outcomes, now the first demonstration of overall survival benefit in perioperative treatment with pembrolizumab with neoadjuvant and adjuvant components of their treatment. I think it’s really important to have a full biomarker panel prior to initiating treatment for any patient who is a candidate for systemic therapy, and in my opinion, those are virtually all patients, and we do see that both immunotherapy approaches, as well as targeted therapy approaches, are moving into the early lines.

We've seen the outcomes of the ALINA study at ESMO [the European Society for Medical Oncology annual meeting] here in Madrid [Spain,] just a couple of days ago, and the benefit is large for an HR of 0.25 for 3 years of adjuvant alectinib. So early biomarker testing is really critical. I think the challenge comes in patients who have issues with turnaround time in the same pragmatic ways as when we’re looking at a metastatic setting. We do want to get patients to treatment. Patients are eager to get treated, but it’s really important to translate the lessons that we learned from the metastatic setting into an early alliance setting.

We talked about PD-L1. I agree with Dr [Kaushal] Parikh, I test all patients, and what we’ve seen is the same correlation between improvements and outcomes between PD-L1 levels in the metastatic setting applied to the early-stage setting. The pCR [pathologic complete response] rate for PD-L1 greater than 50% was 50% in some of the studies that we’ve seen. So very important that they’re not biomarker-agnostic treatments even though they’re biomarker agnostically approved.

So I think this is going to be the message. Liquid biopsy in the early-line setting may fill in, but I think the sensitivity is going to be lower. So I think we’ll still have to optimize our biomarker testing strategies by optimizing tissue utilization but ensuring that we have these ready at the time of decision-making and prior to initiating treatment. There will be a small subset of very early-stage disease that will go straight to surgery, and I think biomarker testing would be a postoperative part, but even in the absence of comparisons of adjuvant vs neoadjuvant approaches, I think the general perception is that neoadjuvant largely supersedes the effect of adjuvant therapies.

I think this follows the melanoma example where we have head-to-head studies comparing those approaches, and it makes sense. We have an intratumoral immunotherapy interaction, and without the tumor microenvironment, it’s not plausible how this would work, and I’m surprised that it does, even though I think it’s much less than we would think of.

Charu Aggarwal, MD, MPH: I think you clearly outlined that it’s very important for us to test for biomarkers PD-L1, EGFR, and ALK in the perioperative, operative, and the adjuvant settings if perioperative or new adjuvant therapy is not being given. Martin, can you walk us through who orders the tests when a patient with stage IV non–small cell lung cancer walks through? Do you have your pulmonologist involved? How do you do it at your institution?

Martin Dietrich, MD, PhD: I think the fault is typically the medical oncologist ordering this. I wish the interventionalists would be involved, but it is, as you know, a fair amount of work getting the tests ordered, and having all the questions from insurance and payers to address. I think that’s something that most interventionalists do not want to take on, so it’s mainly us, but what we’re trying to do, and I think we’ve had some success with local pathologists, [is] to have reflex panels in place that allow us for a faster turnaround time to avoid the administrative delays internally.

I think that's probably going to be a winning strategy. One of the biggest questions was always: What if we test somebody who is inappropriate? I think the message after this, ESMO and ASCO [the American Society of Clinical Oncology annual meeting], is it’s very clear that there is no longer a [patient with] lung cancer who is inappropriate for treatment and independent of stage and histology. We need comprehensive biomarker testing on all lung cancers for our treatment decision-making.

Charu Aggarwal, MD, MPH: I think I would just add that even if some results are not being acted upon prior to the first line, I think we should remember and inform our audience and just remind everyone that there are actionable mutations that can be acted upon, so to speak, in the second-line settings, such as KRAS G12C, HER2, as well as EGFR exon 20. I think just having that information on file is so critical.

Transcript edited for clarity.