EGFR TKIs are the standard of care for the first-line treatment of EGFR mutation-positive NSCLC [6, 7]. In Japan, the phase II, single-arm JO22903 study demonstrated efficacy of erlotinib monotherapy in EGFR mutation-positive NSCLC, with a reported median PFS of 11.8 months . In this updated analysis of the JO22903 study, the 30-month OS rate was 57 % (95 % CI 47–67) and median OS was 36.3 months (95 % CI 29.4–NR). These findings represent a more favorable OS than observed in previous studies of first-line erlotinib in EGFR mutation-positive NSCLC outside Japan (median OS range 22.9–26.3 months [12, 13, 16]), and are in line with results from prospective studies of other EGFR TKIs in Japanese populations (median OS range 27.7–34.8 months [10, 11]). Recently, a median OS of 46.9 months was reported for Japanese patients who received afatinib in the LUX-Lung 3 study , which was longer than that observed in the entire study population . Across these studies, the median PFS values observed in Japanese and global populations were very similar, at approximately 1 year [10–13, 16–18]. Thus, it seems that the current treatment landscape in Japan may be contributing to a longer OS compared with non-Japanese populations, and that OS in Japanese populations can reasonably be expected to reach beyond 3 years.
Although patients with brain metastases have a poor prognosis, which is reflected by the shorter median OS for this subgroup, the findings of this present analysis suggest that erlotinib could be considered effective for patients with brain metastases, as only four patients had CNS progression. This finding is consistent with the phase II ASPIRATION study in Asian patients, which reported that just 4.3 % of patients treated with post-PD erlotinib had new brain lesions . This role for EGFR TKIs has also been observed in populations not restricted to Japanese or Asian patients [20–22]. A case series of 15 patients with NSCLC with EGFR mutations and CNS metastases who received cerebrospinal fluid concentration (CSF) examinations during EGFR TKI treatment provides further evidence to support this conclusion. In this case series, CNS response rate was 57 % with a favorable penetration rate of erlotinib in the CSF . The penetration rate of erlotinib may be dependent on its affinity for p-glycoprotein, which pumps drugs out of the CNS. These findings suggest that erlotinib has a favorable pharmacokinetic profile as a treatment option for patients with brain metastases.
Patients with an exon 19 deletion appeared to have longer OS in our analysis than those with exon 21 L858R EGFR mutation-positive NSCLC. This is similar to the results of a meta-analysis of seven trials (n = 1649), which concluded that patients with an exon 19 deletion had better efficacy outcomes than patients with exon 21 L858R EGFR mutation-positive NSCLC, regardless of which EGFR TKI they received . These data suggest that patients with exon 19 deletion and exon 21 L858R EGFR mutation are clinically distinct populations that should be evaluated further.
In the present study, there was no apparent difference in OS according to subsequent treatments. Median OS was similar for patients who received EGFR TKIs as post-PD therapy (n = 36), which were mainly continuous erlotinib administration following RECIST PD (n = 21) (Table 3), and for those who did not. In contrast to our findings, in a retrospective study of patients with activating EGFR mutations (n = 123) who were treated with EGFR TKIs, OS showed a trend in favor of continuing versus discontinuing EGFR TKI treatment following RECIST PD (33.0 versus 21.2 months, respectively; p = 0.054) . Furthermore, a retrospective clinical modeling study that evaluated the usefulness of EGFR TKI failure pattern for selecting subsequent management, suggested that the efficacy of EGFR TKI continuation differed between patients with gradual progression, local progression, and dramatic progression . Thus, one hypothesis for the inconsistency between studies is the difference in the EGFR TKI failure pattern. Meanwhile, in the present study, various EGFR TKIs were used as post-PD therapy (i.e., erlotinib beyond progression, erlotinib re-challenge after another treatment, and other therapies), which should be noted as one of the limitations. As effective post-PD therapy options are important for patients with disease recurrence, any benefit of EGFR TKI re-administration or continuation after PD requires further study.
At this updated analysis, no new safety signals for erlotinib were observed; single-agent erlotinib was well tolerated and had an acceptable and manageable safety profile in EGFR mutation-positive NSCLC. The safety profile of erlotinib was also in line with previous studies of first-line erlotinib , with the most common AEs being rash and diarrhea.
In conclusion, single-agent erlotinib resulted in a median OS of 36.3 months in the first-line treatment of EGFR mutation-positive NSCLC. Subgroup analyses of OS suggested that the presence of brain metastases was a negative prognostic factor, as these patients had shorter median OS compared with other subgroups. No further differences in OS between specific EGFR subgroups were observed. Although many patients went on to receive additional EGFR TKI therapy following progression, there was no significant difference in median OS for patients who received EGFR TKI as post-PD therapy compared with those who did not. The findings of this single-arm study should be validated in randomized controlled trials.