Mutation Testing for Advanced Lung Cancer

Jared Weiss, MD: To reach standard of care, every stage IV non-squamous patient needs to be tested for molecular characterization. At the minimum, starting with the minimum standard, that needs to include EGFR, ALK, and ROS1. In my practice, we actually expand that out. I think it’s reasonable to expand that out. We know that the overwhelming majority of stage IIIB non—small cell lung cancer will recur. For every non-squamous IIIB patient, I get comprehensive molecular characterization, so that when they become stage IV, I don’t have to wait for that information. I have it available to me. Cure rates for IIIA are also inadequate. We cure about 15% of IIIA non–small cell lung cancer. So, I’m getting molecular testing on these patients, again, because of the high risk of relapse to not wait for the results.

We also have available at my institution, the ALCHEMIST trial. This is a national NCI-led trial for patients with EGFR mutation and ALK gene rearrangement, looking at the appropriate targeted agents in the adjuvant context. And so, for patients who are eligible, which is patients with a tumor at least 4 cm in size or with positive nodes, I’m also getting molecular characterization to be able to consider that trial for them. I also will get testing on patients with other characteristics. So, if there is any non-squamous component to an otherwise squamous cancer—a mixed histology or an adenosquamous histology—in my practice, I’m getting molecular testing. In addition, I’m getting testing on squamous patients if they are never-smokers.

To reach competent practice, in my opinion, at the absolute minimum—and the NCCN Guidelines would endorse this as well—you must get EGFR, ALK, and ROS1. In my clinical practice, I test much more broadly than this. My institution and many others have moved over from à la carte testing with PCR (polymerase chain reaction) and FISH (fluorescence in situ hybridization) to a more broader testing strategy. At my institution, that’s with next-generation sequencing with advanced bioinformatics. Up front, this is much simpler. Instead of checking boxes for what you want, you just check molecular testing and you get a broad characterization of the tumor. Out back, that’s more work. Because if you get something other than EGFR, ALK, or ROS1, you need to interpret it and decide: is it actionable as standard of care? Is it actionable off-label? And if so, should that be done, or is it actionable on clinical trial?

There are a variety of platforms available for molecular testing. At some institutions, à la carte testing is being done for simple gene mutations such as EGFR. This is most commonly done through PCR. For gene rearrangements or gene amplifications, the two FDA-approved actionable ones are EML4-ALK and ROS1. This is accomplished through FISH assay. Many institutions, including my own, are moving over to broader testing platforms. We’re using next-generation sequence with advanced bioinformatics. That’s available in-house at my institution and many others. There are also multiple commercial platforms, with validation behind them, that are quite reasonable.

In order to action a mutation, you need to know that it’s there. And, therefore, molecular testing is required. This is not possible for every patient. Some tumors are anatomically located in a place where it is dangerous to get tissue, or it’s not comfortable for the patient, or the patient declines, or local resources make it very challenging to get the biopsy. And these can be challenging clinical situations where non—tissue-based testing might offer an alternative. One of these, the most dominant, is the plasma-based testing. The idea here is that some tumors secrete their DNA into the blood, and you can therefore test for the presence of these mutations in the blood. Multiple of these tests are commercially available. To my knowledge, only one of them has a nod from the FDA, which is cobas’ EGFR test, but these are pretty broadly available. The rest have different degrees of validation, but they are an option in the clinic right now. Some are using à la carte testing, just as we’ve talked about with tissue, while others are using a broader next-generation type platform.

Weiss case 1:

A 65-year-old man with stage IV NSCLC.

• Tissue biopsy showed EGFR+ adenocarcinoma
• The patient was enrolled in a clinical trial of erlotinib vs erlotinib/bevacizumab
• He was treated with erlotinib for 10 months before developing asymptomatic progression, with slowly growing lesions.
• He was continued on erlotinib for 3 more months; His next scan showed rapid progression.
• Repeat bronchoscopy and mutation testing showed an acquired EGFR T790M mutation
• Patient was subsequently switched to osimertinib