This is the first analysis to reveal the potential implication that afatinib purifies the T790M mutation more than the 1st -G EGFR-TKI. This study suggested that the T790M ratio in 1st or 2nd -G EGFR-TKI-refractory tumors significantly correlated with the response to osimertinib. The T790M ratio in the obtained CR or PR with the osimertinib group (mean: 0.403) was significantly higher than its obtained SD or PD (mean: 0.223); p=0.0272. However, we did not observe a difference in the T790M ratio between the 1st and 2nd G EGFR-TKIs. The T790M ratio in the Afa group (mean: 0.3711) was similar to the 1st-G group (mean: 03616); p=0.9693.
Osimertinib was developed as a drug against T790M by covalently binding to T790M-muted EGFR [10]. Therefore, although osimertinib is also effective against sensitive EGFR mutations, the most effective targets are T790M-positive EGFR mutations. In contrast, afatinib irreversibly binds to EGFR, ERBB2, and ERBB4 and blocks transphosphorylation of ERBB3 to inhibit all ERBB family signaling [20]. Two hypotheses were proposed: the first is that the broader inhibition of the EGFR family may improve the efficacy of EGFR-TKIs for NSCLC with EGFR mutation, and the second is that the broader inhibition of the EGFR family may attenuate T790M cells in cancer cells at relapse after EGFR-TKI treatment. Osimertinib could also be more effective against the AR of T790M after EGFR-TKI. The results of LUX-LUNG 7 and ARCHER 1050 supported this theory in the first-line treatment for NSCLC with EGFR mutation [16] [21]. In the analysis of GIO-TAG trials, the duration of response to osimertinib in patients receiving afatinib and subsequent osimertinib was highly encouraging [17, 22, 23].
In our study, the concentration of T790M correlated with the response to osimertinib, and our results suggested that the degree of attenuation of T790M is important in eliciting the effects of osimertinib. Some prior studies have also shown a similar result: the ratio of T790M to EGFR-activating mutation in cell samples or plasma samples may predict the response to osimertinib or rociletinib [24-26].
However, we did not observe differences in the clonality of T790M after AR between afatinib and 1st -G EGFR-TKI. One reason was that the response of osimertinib after AR was the same in Afa group and 1st-G group when we collected the available tissue, although our previous study revealed that the objective RR and DCR were significantly higher in the Afa group than in the 1st-G group [18]. We intended to collect all the samples in our previous study; however, this was not possible. Another reason is that later lines of therapy may have advanced clonal evolution and may not reflect the effect of osimertinib after 1st-G EGFR-TKI and afatinib treatment.
Next, Pearson correlation analysis showed no significant relationship between T790M ratio and PFS of osimertinib after AR of EGFR-TKIs (r=0.06123, p=0.715). We believed that the PFS of osimertinib after AR in NSCLC with T790M mutation was influenced by co-mutation.
Based on clinal evolution, which was reported for the first time as early as 1976 [27], lung adenocarcinoma seems to follow a multistep progression from atypical adenomatous hyperplasia to adenocarcinoma in situ, and finally invasive adenocarcinoma [28]. EGFR driver alterations are acquired in the early step of cancer progression and can be identified in most neoplastic cancer cells [29]. Furthermore, by layering exacerbations of disease after treatment, especially EGFR-TKI use, cell progression will show different genomic patterns of selection through different lines of therapy [30]. In this setting, EGFR-sensitive mutant tumor cells may coexist with sub-clonal tumor cells with other gene mutations, and co-mutations such as TP53 affect the efficacy of osimertinib [31, 32].
We must consider some limitations in the present study. First, this was a retrospective study, and the participants were selected based on obtainment of adequate samples. Second, the line of treatment for EGFR-TKIs was not determined. Since the lines of EGFR-TKIs varied from case to case, the effectiveness of osimertinib may also vary accordingly. Although osimertinib response was not different between the two groups, we focused on the attenuation of tumor cells by EGFR-TKIs immediately before the administration of osimertinib to clarify the correlation between T790M ratio and osimertinib response.