Roy S. Herbst, MD, PhD

Meeting Insights
Roy S. Herbst, MD, PhD
 

Insights From ESMO Congress 2019: Osimertinib in Non-Small Cell Lung Cancer

Roy S. Herbst, MD, PhD

Practice Community

New Haven, Connecticut

Practice Niche

Lung Cancer

Hospital and Institutional Affiliations

Chief of Medical Oncology, Yale Cancer Center and Smilow Cancer Hospital

Question

In the FLAURA trial, osimertinib provided a statistically significant and clinically meaningful improvement in overall survival vs a comparator EGFR-TKI in the frontline setting in patients with EGFR-mutated advanced non-small cell lung cancer (NSCLC). What is the significance of not requiring magnetic resonance imaging (MRI) upon enrollment of a study like this, and how could that requirement have likely influenced the study findings in FLAURA?

Answer

While MRI was not required, meaning some brain metastases might have been preexisting, I believe the data still support the fact that osimertinib is having an effect.

Question

In your practice/experience, how many lines of tyrosine kinase inhibitor (TKI) therapy do patients with EGFR-mutated disease typically get, and how is this number of therapies likely relevant in the scope of the FLAURA trial?

Answer

In my practice, patients with EGFR-mutated disease typically get 1 line of TKI therapy. Now, there is some thought on giving a second-line TKI if they are resistant with a C797S mutation. This is a big question now: whether to treat with osimertinib like in FLAURA or use a first- or second-generation TKI followed by osimertinib at resistance.

Based on the FLAURA results with the survival advantage and with the median survival improvement of 7 or 8 months, my preference would be – at least in the United States – to continue to use osimertinib first. It is clearly a better drug, a more specific drug, and a less toxic drug. Why would you save your best drug for later?

Question

Researchers found that circulating tumor DNA (ctDNA) monitoring may allow for earlier identification of patients who may progress on first-line EGFR-TKI therapy, and in the detection of EGFR-mediated resistance. What are some limitations of ctDNA analysis in the setting of NSCLC?

Answer

One of the possible limitations is the sensitivity of ctDNA. Is there enough disease for [the test] to pick it up? Certainly if you have metastatic disease and enough blood that the test picks it up, that could provide very early evidence that you should start a new therapy. We are seeing more and more studies using ctDNA in that way. I’m supportive of that but I don’t know that it is a standard of care right now.

Question

Patients with certain mutations may not see an overall survival benefit from osimertinib (ie, Exon21 L858R EGFR mutations), and those with other distinct mutations (ie, TP53 exon 8 mutations) demonstrated primary resistance to frontline osimertinib. Should gene sequencing precede osimertinib administration in the frontline setting, then?

Answer

Gene sequencing prior to osimertinib in the frontline setting is the standard of care. That is how we would know someone has an EGFR mutation. There are currently no known mutations in exon 19 or 21 that would exclude a patient from the use of osimertinib.

Gene sequencing should be used before first-line therapy with an EGFR TKI and before second-line therapy. We should always rebiopsy. There is a whole spectrum of resistance mutations that are emerging. That is what our group at Yale studies as part of our SPORE (Specialized Programs of Research Excellence). With osimertinib, you don’t get T790M mutations but you get a whole different panel of mutations based on that.

Question

What are some comutations of EGFR that you have seen at diagnosis in your patients with lung cancer, and how could these comutations affect treatment selection?

Answer

Some of the commonly occurring comutations with EGFR are C797S or cMET; sometimes the patient develops small cell lung cancer.

Comutations of EGFR would not really affect treatment selection, but you could imagine that we might learn who is more likely to develop which mutation. Someday we might be able to use that in our treatment selection decisions, but right now we don’t use that information.