While this study indicates that approximately of patients
While this study indicates that approximately 10% of patients treated with afatinib are likely to be LTRs, we were unable to identify any baseline clinical or tumor characteristics that could potentially identify patients most likely to achieve long-term benefit. The baseline characteristics of afatinib LTRs were generally comparable with those of the overall afatinib-treated groups in the LUX-Lung 3, LUX-Lung 6, and LUX-Lung 7 studies. The frequency of Del19 mutation-positive NSCLC was slightly higher among LTRs, presumably reflecting its known status as an indicator of improved outcomes with EGFR TKIs vs. other mutation types [29,30]. Importantly, other factors, such as presence of CAY10683 metastases or uncommon mutations, did not preclude long-term response to afatinib. These findings are consistent with previous analyses of LUX-Lung 3 and 6 that showed that afatinib can provide clinical benefit in advanced NSCLC patients with brain metastases or uncommon mutations [31,32]. In our analysis, the proportion of LTRs with brain metastases at enrolment was slightly less than in the overall study populations. Nevertheless, some patients with brain metastases at baseline (across all 3 trials) were able to obtain long-term clinical benefit with afatinib. Furthermore, previous analyses of patients with tumors harboring uncommon mutations in LUX-Lung 3 and 6 showed that afatinib demonstrated activity in certain uncommon mutation types . We also noted that LTRs in LUX-Lung 3 and 6 included patients with NSCLC harboring uncommon mutations (specifically, S768I, or G719X mutations). Our analysis of molecular factors predictive of long-term response is limited because tumor tissue or plasma samples were not collected at baseline or at the time of progression. As such, it was not possible to identify potential molecular biomarkers indicative of long-term response to afatinib. Safety analysis in this post-hoc analysis focused on AEs that were most frequently reported in afatinib-treated patients in the LUX-Lung 3, LUX-Lung 6, and LUX-Lung 7 trials (diarrhea, rash/acne, and stomatitis). The incidence and onset of these AEs among LTRs was consistent with previously reported findings for the overall afatinib treatment groups in LUX-Lung 3, LUX-Lung 6, and LUX-Lung 7 [, , ]. Since only selected AEs were evaluated, we cannot conclude from this analysis whether other safety signals were more or less frequent in LTRs. Afatinib-related AEs have previously been shown to be effectively managed with supportive care and tolerability-guided dose adjustment . The frequency of dose reductions with afatinib due to treatment-related AEs was similar between the LTRs and the overall afatinib treatment groups in LUX-Lung 3, LUX-Lung 6, and LUX-Lung 7 [, , ]. Among the LTRs to afatinib, dose reductions tended to occur early in the treatment course (most within 6 months of initiating treatment). Previous analyses have shown that tolerability-guided dose adjustment of afatinib does not impact efficacy  and our analysis indicates that tolerability-guided dose adjustment does not prevent a patient from having a long-term response. Indeed, some LTRs had their dose reduced to 20 mg shortly after initiation of treatment but obtained long-term clinical benefit with afatinib. This highlights the importance of determining a tolerable dose for each patient on a case-by-case basis. The long-term tolerability of afatinib was further substantiated by assessment of PROs in afatinib LTRs. PRO scores in afatinib-treated LTRs remained stable, and were slightly improved after approximately 3 years of treatment compared with the scores at treatment initiation. Again, this is consistent with data from the overall populations of LUX-Lung 3, LUX-Lung 6, and LUX-Lung 7, which showed that afatinib was associated with clinically relevant improvements in QoL and lung cancer symptoms [, , ]. In the current analysis of LUX-Lung 3, LUX-Lung 6, and LUX-Lung 7, long-term treatment with afatinib had no detrimental impact on the use of subsequent treatments. As in the overall populations of LUX-Lung 3, LUX-Lung 6, and LUX-Lung 7, patients were able to receive a variety of subsequent treatment options, including chemotherapy and/or other EGFR TKIs. The rates of subsequent therapy uptake were lowest in LUX-Lung 6 (29%), which may be due to reimbursement and availability of specific drugs in the countries involved in the study. Duration of subsequent treatment also appeared similar to the overall LUX-Lung 3, LUX-Lung 6, and LUX-Lung 7 populations. When looking at subsequent treatments in the individual trials, it is of note that most afatinib LTRs in LUX-Lung 7 (74%) were still on afatinib treatment at the time of analysis, so have not yet received any subsequent treatment. Following recent progress in our understanding of resistance mechanisms to afatinib, it is now clear that the emergence of the T790M mutation is as common in afatinib-treated patients (50–70%) as it is in those receiving erlotinib or gefitinib . In principal, therefore, the majority of LTRs to first-line afatinib would be fit enough, and eligible for, subsequent treatment with osimertinib. In this study, few LTRs had received osimertinib. This reflects the fact it has only recently been approved in this indication, based on the AURA and AURA3 trials [34,35].