The focus of the presentation was a description of acquired resistance mechanisms to osimertinib when it is given as second line therapy for EGFR mutated non-small cell lung cancer. The results were obtained from a phase III clinical trial, AURA 3, that randomised patients to osimertinib or platinum pemetrexed chemotherapy after receiving first line EGFR tyrosine kinase inhibitor and demonstrating progression and after confirming centrally that the tumour harboured a T790M mutation. The analysis was derived from plasma samples that were obtained at baseline and at discontinuation or progression and the focus was description of acquired resistance mechanisms.

The highlights from the presentation: about half of the patients have lost T790M in the progression or discontinuation specimen, something that has been previously described, and about 17-19% develop MET amplification. About 11% of the patients have developed a new EGFR mutation, C797S, and a number of other alterations, most of which have been previously described in smaller studies were seen, such as gene fusions or amplification of cell cycle genes, amplification of PI3 kinase.

So the study serves as the largest cohort of patients in the second line osimertinib exposure and most likely a reference study for the future. The other interesting part about the study was that it was presented side by side with results regarding acquired resistance when osimertinib is given as a first line therapy. There were similarities between the mechanisms of resistance between the two groups and some differences. For example, if osimertinib is given as first line therapy we almost never see T790M emerging. Also resistance mechanisms described in first line appear to be a little less frequent than they were in the first line therapy but again they are dominated by MET amplification and alternative EGFR mutations.

What do you think these differences between first and second line might reveal about the tumour evolution that is driving the lung cancer?

Certainly, also because there is selective pressure on clones that exist in second line therapy patients, namely T790M, which in about half of the patients disappears from the specimen. In first line patients T790M is not expected to be there so we hardly see any events pertaining to T790M in the front line.

Will the lessons from this trial be useful for other clinicians going forwards in identifying the stages of relapse for any patients in their clinic for finding out who is best treated for what?

I think the major utility of these findings is in guiding our research in therapeutic interventions. It is increasingly clear that there will be no one answer fits all for patients that develop resistance and therefore we will have to identify the mechanism of resistance to identify also the targeted therapy or other therapy that is suitable for the patient.