Review Girard Third-generation EGFR TKIs as first-line treatment Osimertinib has recently been evaluated in a first-line setting in the randomized Phase III trial, FLAURA [10] . In this study, patients with common (Del19 / L858R) EGFR mutations were randomized to osimertinib or a first- generation EGFR TKI, whereby the specific EGFR TKI (erlotinib or gefitinib) was chosen by the site as the sole comparator prior to site initiation. The primary end point – PFS by investigator assessment – was significantly improved in the osimertinib arm versus the erlotinib / gefitinib arm (median, 18.9 vs 10.2 months; HR: 0.46; 95% CI: 0.37–0.57; p < 0.0001; Table 1 ). PFS benefit was consistent across subgroups, including race, age, gender and presence of stable CNS metastases at baseline. The response rate was similar between treatment arms (80% with osimertinib and 76% with erlotinib / gefitinib; OR: 1.28; 95% CI: 0.85–1.93; p = 0.2335). However, median duration of response was 17.2 months with osimertinib versus 8.5 months with erlotinib / gefitinib. At data cutoff, OS data were immature (25%). Importantly, censoring of OS started at 15 months, which is when crossover from erlotinib / gefitinib to osimertinib may start to have an impact [10] ; as such, the OS results are awaited with interest. Similar to the results from AURA and AURA3, plasma testing for EGFR mutations within the FLAURA trial showed good concordance with tissue testing, supporting the use of plasma testing in clinical practice to identify patients suitable for treatment [42] . While FLAURA supports that osimertinib is more effective than erlotinib / gefitinib, the results should be considered further in light of several limitations. First, despite being the only approved second-generation EGFR TKI, afatinib was not included as a comparator. In light of emerging differences between first- and second- generation EGFR TKIs observed in LUX-Lung 7 and ARCHER 1050, this precludes any evaluation of clinical benefit for osimertinib over a second-generation EGFR TKI. Additionally, as well as the OS data from FLAURA (and AURA3 for the sequencing strategy), we await further details on efficacy by TKI comparator (erlotinib or gefitinib) and information on subsequent therapies. Based on current standard practice, most patients with EGFR mutation-positive NSCLC receive a first- or second-generation TKI in the first-line until disease progression. Will results from FLAURA change this treatment paradigm? While it would appear likely that osimertinib will become a first-line treatment option for EGFR mutation-positive NSCLC, the optimal first-line choice remains complex. In the next section, we consider some of the key arguments that have emerged following presentation of the FLAURA data. Factors influencing first-line treatment choice: mechanisms of resistance & subsequent therapy options With the availability of multiple agents for EGFR mutation-positive NSCLC, treatment needs to be considered in terms of a long-term plan to maximize survival. Thus, the sequence of treatment and consideration of the molecular status of the tumor is of key importance. In the Phase III trials of first-generation EGFR TKIs, rates of subsequent therapy were generally high (around 60–70%) [17–18 , 20] . With regard to the second-generation EGFR TKIs, detailed analysis of the LUX-Lung 3, 6 and 7 trials showed that 71% of patients received a further line of treatment following discontinuation of afatinib, with 28% of patients receiving four lines of subsequent treatment [43] . Subsequent therapy after afatinib predominantly consisted of platinum-based chemotherapy, with others including single-agent chemotherapy, first-generation EGFR TKI monotherapy, chemotherapy combination therapies and third-generation EGFR TKIs, thus suggesting a range of potential treatment options post-afatinib that may impact survival of patients. Furthermore, subsequent treatment duration was similar irrespective of EGFR mutational subgroup (Del19 / L858R) [43] . As previously discussed, acquisition of an EGFR T790M mutation is the predominant mechanism of resistance for first- and second-generation TKIs. For patients with T790M-positive NSCLC who have progressed on an EGFR TKI, osimertinib is now a clear treatment of choice based on data from AURA and AURA3. Due to lack of availability of third-generation EGFR TKIs and the fact that T790M testing was not mandated at the time the LUX-Lung 3, 6 and 7 trials were conducted, only a minority of patients (n = 37) received osimertinib, primarily in the third- or later-line setting [43] . Analysis of these patients has to be interpreted with caution, based on small sample size and the possibility of sample enrichment; for instance, the median duration of treatment with first- line treatment in these patients was notably long at 21.9 months. Nevertheless, exploratory hypothesis-generating analysis of time on osimertinib and OS in patients starting on afatinib who received subsequent osimertinib in any line was highly encouraging. Median time on osimertinib in any treatment line was 20.2 months, and median follow-up for osimertinib-treated patients was greater than 4 years. At this time, median OS had not yet been reached [43] . In the current treatment landscape, it is likely that many patients will be eligible for, and able to receive 10.2217/fon-2017-0636 Future Oncol. (Epub ahead of print) future science group