Wade T. Iams, MD, discusses molecular testing and multidisciplinary treatment for a patient with non–small cell lung cancer.
A White man aged 66 years presented to his primary care physician complaining of visual disturbances, fatigue, and sporadic headaches. He had a medical history of hypertension managed on candesartan; hyperlipidemia managed on simvastatin; and a 25-pack-year smoking history. A physical examination showed blood pressure at 148/70 mm Hg and decreased breath sounds in the lower left lobe, but it was otherwise negative. Complete blood count and chemistries were within normal limits.
A brain MRI demonstrated a 10-mm right parietal mass at the gray-white junction without significant vasogenic edema. A CT scan of the chest, abdomen, and pelvis revealed a 3.4-cm mass in the lower left lobe and several small liver metastases. Ultrasound-guided core needle biopsy of the lung lesion showed grade 2 lung adenocarcinoma, acinar subtype. Staging was T2aN0M1c, and his ECOG performance status was 1.
The patient had 90% PD-L1 expression on tumor cells (22C3 pharmDx test) and was anxious to start treatment. Molecular panel testing showed EGFR exon 19 deletion.
Regarding non–small cell lung cancer (NSCLC) molecular testing in your practice/institution:
IAMS: Does anyone want to comment on whether molecular testing is ordered for every patient with metastatic lung adenocarcinoma?
MAHESHWARI: We [at Owensboro Health Medical Group Hematology and Oncology] order it for every patient.
BRENNAN: [At St Elizabeth Healthcare], we do not. The original studies on biomarker-based therapies [involved patients with] ECOG performance status of 0, 1, or 2. If I have somebody with a bad performance status of 3 or 4, I don’t test them. They’re not candidates for therapy, probably 10% or 15% of patients.
RESKE: I think you ought to at least see the patient and then get a sense of whether they’re interested in treatment, which the majority of patients are. But you do run into patients who are not interested in treatment, so why send all this work-up?
IAMS: That is absolutely an important point. We [at Singing River Health System] check every patient’s tissue, as well as their peripheral blood.
ANTHONY: One of the reasons [University of Kentucky Markey Cancer Center] can’t get reflex testing is that the pathologist is not sure the patient wants treatment.
IAMS: It sounds like when patients seem interested in treatment, if the performance status permits, then everybody [gets tested]. I agree, those are all important considerations in our processes.
For the next question, we haven’t surveyed the breadth of the panels [being done]. For example, we’re not doing a targeted panel. It’s just a comprehensive panel, several hundred genes initially from the tissue.
Does anyone want to speak to [whether] you’re doing targeted panels in patients with metastatic lung adenocarcinoma and if so, which genes do you start with?
BRENNAN: We just order a next-generation sequencing [NGS] assay and get the whole [list].
IAMS: Is anybody doing rapid EGFR or ALK testing? Some institutions are doing that, more though in the resectable setting if neoadjuvant chemoimmunotherapy is being considered. But I think [today], ordering a broad NGS is the best approach.
Is everyone doing PD-L1 testing? If so, how is that helping your treatment decision?
BALA: Yes, I do order PD-L1 testing for all patients. I can’t say it’s helpful unless I’m considering single-agent pembrolizumab [Keytruda]. If the patient is not a candidate for chemotherapy or if the PD-L1 expression is high, then it helps me to use pembrolizumab. But [because] the chemoimmunotherapy combination can be used regardless of PD-L1 expression, it’s not helpful in every patient.
MAHESHWARI: We do [PD-L1 testing for] every patient. That’s required by the insurance now because they don’t approve any patient’s immunotherapy drugs in the patients with stage IV disease unless you have this molecular testing.
IAMS: I have intermittently run into that insurance issue, particularly where we’ve been seeking to start treatment rapidly with chemoimmunotherapy. It is approved across PD-L1, but I’ve encountered that issue. But [for those with] PD-L1 greater than 50%, pembrolizumab monotherapy is a big reason we’re still doing the test in every patient.
It’s important to note that benefit of pembrolizumab monotherapy continues to be stepwise even above PD-L1 expression of 50%.1 So for PD-L1 of 90%, the odds of response and durable benefit [are] higher than 70% PD-L1, for example, but that nuance is only in a subset of patients. Our studies would suggest that only about [one-third] of patients have PDL1 greater than 50%, but I think it can be helpful, particularly as noted by Dr Bala, if we’re considering pembrolizumab monotherapy.
• For the patient in this case with NSCLC, would you initiate multidisciplinary assessment for treatment of his central nervous system involvement? If so, would you opt for stereotactic radiosurgery (SRS) before instituting the tyrosine kinase inhibitor (TKI)?
• Do you prefer to use osimertinib (Tagrisso) up front or defer its use in case an EGFR T790M mutation develops on first-generation TKIs?
IAMS: The patient has EGFR exon 19 deletion, brain metastases, multiple liver metastases, and a lung lesion. Do you talk about it with your radiation oncologist? If there are neuro-oncologist consultants there, do you involve them? How do you think about those patients with de novo brain metastases and targetable EGFR mutations and how do you approach it in the clinic?
RESKE: I think it’s worthwhile to get different opinions. When I refer to a radiation oncologist, sometimes it’s not necessarily perceived as answering the question whether it’s a good idea to radiate…. That answer might be different when you send someone to a neuro- oncology colleague. That’s at least…my perception. It depends on what radiation oncologist you send the patient to.
IAMS: That’s exactly my experience and intention. There is a lot [based] on whomever is seeing them at the initial diagnosis and identification of these potential targeted mutations. In my experience, it’s the same. Most of the radiation oncologists [whom] we collaborate with will usually treat the patient if they’re referred to them. I think it’s complex.
There are a couple of neuro-oncologists [whom] I’ll often involve and have them comment on the safety of attempting treatment with a TKI before SRS on a lesion like this. But it’s always a tough debate.
MAHESHWARI: I’ve seen multiple patients where we give carboplatin/ pemetrexed-based treatment in those with adenocarcinoma and brain metastases. The brain lesions melt away completely. I have 4 or 5 patients with stage IV disease who are 5 years out. So, in a case like this…we start treatment anyway because we are waiting for the results. Why can’t we complete 2 or 3 cycles, then change the treatment?
IAMS: The National Comprehensive Cancer Network guidelines changed as far as if systemic therapy is initiated and a targetable oncogene is found— the debate between continuation to maximal benefit with that initial systemic therapy vs changing to the targeted agent.2 The data sets that I think about and the way I approach it are that targetable oncogenes are better associated with never-smokers or very little smoking and low tumor mutation burden like EGFR, ALK, ROS1, RET, or HER2 mutations.
Those are the mutations where we have some data sets commenting on the least likely effective treatment with immune therapy. Our other potential targets like MET exon 14, KRAS G12C, and BRAF mutations have more data that those patients may benefit from immune therapy, potentially similar to patients with wild-type mutations. Then there are some data sets in the KRAS mutation setting, sometimes [showing] more benefit with immune therapy.
So, in those settings, I think continuation to maximal benefit is reasonable. I think with the low tumor mutational burden targetable mutations, I would often change it to the targeted therapy there, but it is a tough question.
Is anyone deferring osimertinib up front to T790M development, starting with first-generation TKIs, or are most individuals starting with osimertinib?
BRENNAN: I start with osimertinib. There weren’t many data [saying] that delaying it is of any benefit.
CHEDID: There were data where therapy with afatinib followed by osimertinib had a longer disease-free interval or longer conversion to chemotherapy interval than going to osimertinib initially.3 But I start with osimertinib.
IAMS: Osimertinib is the preferred initial treatment for patients with canonical EGFR mutations, so L858R and exon 19 deletion.2