Pennell Shares Best Practices for Molecular Testing in Lung Cancer

Nathan Pennell, MD, PhD

In the scenario where symptomatic patients with lung cancer require immediate treatment but have no molecular testing results, Nathan Pennell, MD, PhD, recommends that oncologists potentially hold off on immunotherapy until all oncogenic drivers have been ruled out.

“Molecular testing upfront as quickly and efficiently as possible is key in every patient with nonsquamous non—small cell lung cancer,” said Pennell. “In the common scenario where you can't get it done quickly and you have to start a patient's treatment, I would not start immunotherapy upfront until you're able to get that testing done.”

Administering immunotherapy to symptomatic patients with lung cancer who require immediate therapy but have no molecular testing results could harm the patient if they have a targetable mutation. Therefore, forgoing immunotherapy for these patients and administering chemotherapy if the patient is unsure of their molecular profile status is an optimal approach, he said.

“The downstream consequence [of not molecular testing] is the patient may miss out on a very effective therapy if you never identify whether they have a mutation. If they have a mutation and they are treated with immunotherapy, they may have significant toxicity and you have to switch them to a targeted therapy,” Pennell added. “Ultimately, try not to get yourself in that situation.”

In an interview with OncLive, Pennell, an associate professor and director of the Lung Cancer Medical Oncology program at Cleveland Clinic, offered strategies regarding molecular testing in patients with lung cancer and how to address treatment if patients are symptomatic and molecular testing results are unavailable.

OncLive: How should you treat symptomatic patients diagnosed with stage III or IV lung cancer who have not received molecular testing but require immediate treatment?

Pennell: When patients are diagnosed with lung cancer, you may have PD-L1 results but not molecular testing results yet because that typically takes significantly longer. Sometimes you don't even have enough tissue left to do the testing. You have to decide whether to start the treatment now, wait, or get more tissue to do molecular testing. This is a very practical issue. You should do your best to not end up in this situation by making sure that you work with your institution to have enough tissue to get proper molecular testing. Try to get it done as quickly as possible by getting your processes as streamlined as possible. You can incorporate plasma testing if you need to get faster results.

Unfortunately, it's still going to happen. You're going to end up having a patient, perhaps with a high level of PD-L1 expression, with no molecular testing. The patient may be very symptomatic, meaning they need immediate treatment. What do you start with? The consequence of missing an oncogenic driver is very significant; the patient may miss out on potentially significantly life-extending therapy with drugs like osimertinib (Tagrisso) if they have an EGFR mutation an ALK rearrangement. Further, if you're giving a patient with an oncogenic driver immunotherapy, they are much less likely to benefit from it compared with patients without driver mutations. The immune checkpoint inhibitor trials showed that mutation-positive patients typically do not benefit from immunotherapy the same way that wild-type patients do.

These drugs stay in a patient’s system for weeks and months. If you later discover that the patient has a mutation, this can cause some consequences. For example, what happens if you start a patient on chemotherapy plus pembrolizumab (Keytruda) and then you discover that they have an EGFR mutation? You could switch them over to osimertinib if they are EGFR-mutation positive, but a phase I/II trial combining osimertinib with an immune checkpoint inhibitor durvalumab (Imfinzi) suggested that the osimertinib/durvalumab combination causes an extremely high rate of interstitial lung disease (ILD). In the dose-expansion phase of that trial, 67% of the patients developed ILD and had to be halted.

In the ALK-positive NSCLC trials, there have been numerous phase I trials. If you have ALK-positive patient, combining checkpoint inhibitors without TKIs has significant hepatic toxicity. As many as one-third to half of patients may have significant hepatic toxicity. You should try to avoid giving a checkpoint inhibitor if you don't have the mutation testing back. You can start chemotherapy alone and wait for the testing. You can always add in the checkpoint inhibitor later on.

What are some common challenges you face without the use of molecular testing?

Even in a high PD-L1—expressing patient, I would not treat someone without molecular results with single-agent immune checkpoint inhibitors. A phase II trial of EGFR-mutant patients showed that patients are extremely unlikely to respond to immunotherapy. In that trial, when they crossed over to having a TKI, 2 of the 7 people who crossed over had either grade 3 hepatic toxicity and 1 patient died from ILD caused by the combination. [Immunotherapy] is actually a fairly significant risk in [EGFR-mutant patients]. Generally, you can give chemotherapy if you have to start [treatment immediately] and wait to get molecular testing results.

One common scenario is patients diagnosed with stage III lung cancer who get chemoradiation, and then consolidation durvalumab for up to 1 year afterwards. In the earlier-stage setting, many people don't do molecular testing. It's not necessarily wrong to not test someone in the stage III setting, because it does not change how you treat them. However, 6 months into durvalumab, when they progress, molecular testing is performed to discover that they have a mutation.

Now you're stuck, because there is a significant risk of putting that person on a TKI. I often recommend starting that patient on chemotherapy alone and giving them at least 3 months to clear [durvalumab] out of their system before putting them on a TKI. There are no prospective trials in this scenario, but the consequences of treating a patient [with immunotherapy] is you may have to stop their treatment and now they do not have any good therapy. They may have a life-threatening complication due to that treatment.

How do liquid biopsies play a role in these situations?

Plasma testing for genetic variations can be incredibly helpful in the common scenario of a patient who doesn't have enough tissue for testing, or you have to wait a significant amount of time [to get the results]. It is very common in my practice for patients to have a prior biopsy somewhere else and we do not have access to their tissue immediately.

Rather, we have to somehow track [the tissue] down and have it sent to us. The pathology department has to send out for a tissue block, and then it has to be tested. It may take a number of weeks or a month to get those results. The patient does not want to wait that long or maybe they are sick and need immediate treatment. That's a setting where liquid biopsies can be extremely helpful.

By itself, plasma testing can't replace tissue testing. It's extremely reliable if you find a positive result of a mutation or a gene fusion on plasma testing. You can use that and go right on to treatment. If a patient has a common mutation like KRAS mutations, they likely don't have another actionable mutation and you can move on to common systemic therapy chemotherapy or immunotherapy. Plasma testing is probably going to miss about 20% of actual mutations, which means if you get a negative result on a plasma test, you have to still follow through and do the tissue testing by either getting a new biopsy or tracking down that tissue. That's a scenario where, if you have to start treatment again, I would usually start with chemotherapy alone and not add in a checkpoint inhibitor until you know for certain that that tissue is negative.