MYLUNG Consortium Finds Molecular Testing Gaps in NSCLC

Results from the first phase of the MYLUNG Consortium research study in metastatic non-small cell lung cancer (NSCLC) were released by the U.S. Oncology Network, U.S. Oncology Research, and Ontada at the 2021 virtual American Society of Clinical Oncology (ASCO) annual meeting. The findings revealed significant gaps in biomarker testing.

In this exclusive MedPage Today video, H. Jack West, MD, a thoracic oncology specialist at City of Hope in Duarte, California, explains that these first-phase results indicate there is still a lot of work to do.

Following is a transcript of his remarks:

Hi, I’m Dr. Jack West, and I’m an associate clinical professor in medical oncology at the City of Hope Comprehensive Cancer Center in the Los Angeles area. I wanted to talk for a few minutes about one of the more influential and very timely presentations from ASCO 2021 in the lung cancer session. And this was by doctors [Nicholas J.] Robert and colleagues. It was actually presented by Dr. [Makenzi] Evangelist at the meeting itself.

And it looked at the biomarker tissue testing process in almost 3,500 patients within the U.S. Oncology Network from spring of 2018 through spring of 2020, over a 2-year period. The group was called the MYLUNG Consortium and looked at testing rates for four tumor markers or molecular markers that were standard of care and included in the NCCN [National Comprehensive Cancer Network] guidelines, having FDA approved targeted therapies associated with them — namely EGFR, ALK, BRAF, and ROS1 — as well as looking at PD-L1 rates and looking at how frequently each of these were tested for in the broader non-small cell lung cancer population, advanced non-small cell, as well as a squamous or non-squamous, versus this broader population.

They looked at next-generation sequencing [NGS]. So broad molecular testing and a bit on turnaround times and the time to get the results in practice.

The key findings were that in the broad population of all non-small cell that they looked at, the rates of getting all five of the biomarkers that were considered relevant was under 50% — 46% specifically. And even when they eliminated patients with squamous histology and just focused on non-squamous, it improved from 46% to 49%, but still less than half the patients had all of these tests done.

Now, 91% of the patients with non-squamous histology had at least one test, but that included PD-L1, which is not a driver mutation test. That was the most common testing done, at 83% in both the broader population and non-squamous.

And among the driver mutations being tested, they topped out at about three out of four for EGFR and ALK in the non-squamous population, 70% of the broader population, and BRAF was only tested in 59% of the non-squamous population — a little less than that, 55% in the broad population.

They also noted that about 10% of the patients had biomarker testing results come back after the start of treatment. And that’s sometimes born out of necessity, but is less than optimal because it really helps us define what the best treatment is. And if someone happens to have a driver mutation, we really would strongly prefer for that targeted therapy to be initiated before they start certainly immunotherapy, but ideally also even before chemotherapy and right out of the gate.

They looked at the changes over the 2-year period in blocks of 6 months, and really did not see any change in the rates of testing over that 2-year interval, with the exception of BRAF testing, which improved from 54% in the first 6 months to 62% by the end.

So I would say that’s kind of damning with faint praise. And the rate of all five being tested, improved from 44% to 52% in that interval. So some improvement, but really not a lot of change over 2 years, during which the momentum of targeted therapies really increased. And when they looked at the rates of NGS testing, it improved from 33% to 45%.

So yes, it was getting better over time, but I would say that this is one of those half-full rather than half-empty situations when we have many relevant targets and still less than half of the patients are getting NGS testing.

One thing they found was that it was about 35 days from the start from the diagnosis to the initiation of treatment. And it typically took 10 to 15 days for biomarker results to come in from the time that orders were put in. And that is looking at all testing methods. It would probably be a good bit higher for NGS testing. And I think that is a limitation of this.

So overall, what can we say about this? One point is that the investigators should be commended for looking into this very important question. The results overall, I would say, are disappointing, because advanced lung cancer is kind of the tip of the spear, the poster child for the value of molecular testing and a setting where we have now a very rapidly growing collection of relevant markers with approved therapies, including several since the time of this retrospective look at molecular testing rates here.

We had MET and RET added in the spring of last year and KRAS just added. So this collection of relevant markers just keeps growing, and that would make NGS testing only more compelling to get all of these and some emerging ones for the future at the same time.

So I think what this really shows us is that there’s still a lot of work to do in assessing what those barriers in real-world practice in the community really are. Because this was a good representative sample, almost 3,500 patients treated in the U.S. Oncology Network, and trying to answer what is the real-world practice patterns and finding that it’s pretty disappointing compared to what we might idealize or envision based on our enthusiasm for molecular testing.

We’re only going to realize the benefits of precision medicine, precision oncology, when far more than 50% and ideally much closer to 100% of patients are getting good, broad molecular testing to identify all of the targets that are relevant today and in the coming years.