Introduction

Treatment with epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors has been shown to improve the prognosis in patients with EGFR-mutant non-small cell lung cancer (NSCLC). At first, first-generation EGFR-TKIs were shown to prolong the progression-free survival (PFS) as compared to cytotoxic agents [1, 2]. Thereafter, afatinib, a second-generation EGFR-TKI, was shown to prolong the PFS than the first-generation EGFR-TKIs [3], and this drug yielded a longer overall survival (OS), although the difference did not reach statistical significance [4]. In a more recent trial, the PFS and OS were significantly prolonged in the osimertinib treatment arm (a third-generation EGFR-TKI) as compared to the control arm treated with a first-generation EGFR-TKI [5]. In addition to these therapeutic advances, reports have shown that about a half of the patients who are treated with a first- or second-generation EGFR-TKI develop acquired resistance to the primary EGFR-TKI therapy, associated with the emergence of the EGFR exon 20 T790M mutation [6], but that the PFS in these patients can still be prolonged by sequential therapy with osimertinib [7]. Other studies have reported that patients who developed resistance associated with the emergence of the T790M mutation after treatment with afatinib who received sequential osimertinib therapy showed an extremely prolonged OS [8, 9].

Based on the above, we conducted a retrospective study to investigate the associations between clinical factors and the OS in patients with EGFR-mutant NSCLC.

Methods

Patient selection

Clinical information was retrieved from the medical charts. The inclusion criteria were as follows: (1) patients with cytologically or histologically diagnosed EGFR-mutant NSCLC; (2) patients with advanced, locally advanced, or relapsed NSCLC who were not ineligible for local therapy; and (3) patients who had received EGFR-TKI therapy as the first-line treatment between 2007 and 2020 at our institution. The exclusion criteria were as follows: (1) patients in whom the exon 20 T790M mutation was detected in the tumor even prior to the start of the first-line treatment.

This retrospective study was conducted in accordance with the principles outlined in the Declaration of Helsinki and Ethics Guidelines for Medical and Health Research Involving Human Subjects (Ministry of Health, Labour and Welfare, Japan). We disclosed information about the study to the patients under the approval of the Ethics Committee, University of Toyama (approval number: R2019103).

Clinical information

The tumor EGFR mutations were classified as exon 19 deletion, exon 21 L858R, or uncommon mutations. Compound mutations with common (exon 19 deletion or exon 21 L858R) and uncommon mutations were classified as uncommon mutations.

Tumor PD-L1 expression was evaluated by immunohistochemistry in formalin-fixed and paraffin-embedded tumor tissue specimens, using the 22C3 antibody (BML, Tokyo, Japan). The proportion of PD-L1-positive cells in the tumor was evaluated as the tumor proportion score (TPS) for PD-L1 expression.

Local therapy for brain metastasis was defined as local therapy administered at the start of first-line treatment and included stereotactic radiotherapy, whole-brain irradiation, and surgery.

Sequential therapy with osimertinib was defined as treatment with osimertinib for patients who developed acquired resistance to the primary EGFR-TKI therapy associated with the appearance of the T790M mutation in the tumor. Re-administration of first- or second-generation EGFR-TKIs was defined as re-administration of other first- or second-generation EGFR-TKIs for patients with acquired resistance to the primary EGFR-TKIs. Changes of the EGFR-TKIs administered for other reasons such as development of adverse events associated with the initial EGFR-TKIs were not included in the above definition.

Statistical analysis

Survival curves were drawn by the Kaplan-Meier method. OS was calculated from the date of initiation of treatment with an EGFR-TKI until death and censored at the last visit of the patient. Patient data were dichotomized according to categorical variables or median values of numerical variables, and the OS was compared by the log-rank test. Variables that were determined as showing an association with P values of <0.2 were selected as independent variables to be entered into the Cox proportional hazards model. All the statistical analyses were performed using JMP ver. 15.0.0 (SAS, Cary, NC).

Results

A total of 158 patients with EGFR-mutant NSCLC were treated at our institution between 2007 and 2020. Of the 158 patients, 25 patients were excluded from this analysis because they did not receive first-line EGFR-TKI treatment, and 3 patients were excluded because the T790M mutation was detected in the tumor even at initial diagnosis. Data of the remaining 130 patients with EGFR-mutant NSCLC who had been treated with EGFR-TKIs as the first-line treatment were included in this analysis. The date of death could be confirmed for 52 (40%) patients, but not for 48 patients (36.9%), because the latter were transferred to some other hospital prior to their death. Anticancer therapy was ongoing at the time of the analysis in this study in 30 (23.1%) patients.

Table 1 shows the patient characteristics. Of the 130 patients, 84 (64.6%), 25 (19.2%), and 21 (16.2%) patients received first-line treatment with a first-generation (including gefitinib and erlotinib), second-generation (afatinib), and third-generation (osimertinib) EGFR-TKI, respectively. Five (3.8%) patients received combined EGFR-TKI plus vascular endothelial growth factor (VEGF) inhibitor therapy, and 36 (27.7%) patients received platinum-doublet therapy at some point during the clinical course. Twenty-one (16.2%) patients received sequential therapy with osimertinib, and 16 (12.3%) patients received re-administration of another first- or second-generation EGFR-TKI. Out of the total of 109/130 patients who did not receive sequential therapy with osimertinib, the anticancer therapy had been discontinued in 43 patients even prior to 2016, when osimertinib therapy first became available. Brain metastasis was detected at the initial diagnosis in 37 (28.5%) patients and was treated by stereotactic radiotherapy in 14 patients, stereotactic radiotherapy and surgery in 1 patient, and whole-brain irradiation in 9 patients.

Table 1 Patient characteristics
Full size table

Table 2 shows the association between the patient characteristics and the OS as determined by the log-rank test; comparison by the log-rank test identified exon 19 deletion in the tumor, Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0–1, a serum lactate dehydrogenase (LDH) level of <200 U/L, and sequential therapy with osimertinib as being associated with a prolonged OS. Figure 1 shows the Kaplan-Meier curves for OS in patients who did and did not receive sequential therapy with osimertinib, excluding patients receiving first-line treatment with osimertinib. The OS duration was longer (median, 78.5 months) in patients who had received sequential therapy with osimertinib as compared to those who had received re-administration of first- or second-generation EGFR-TKIs (median, 31.5 months) and patients who had received neither of the above (median, 32.3 months) (P = 0.013, log-rank test).

Table 2 Associations of the patient characteristics with the OS (log-rank test)
Full size table
Fig. 1
figure 1

Kaplan-Meier curve for OS in patients who received sequential therapy with osimertinib, who received re-administration of the first-/second-generation EGFR-TKIs, and who received neither of the above. Solid line: the patient group that received sequential therapy with osimertinib; dotted line: the patient group that received re-administration of first-/second-generation EGFR-TKIs; dashed line: patients who received neither sequential therapy with osimertinib nor re-administration of first-/second-generation EGFR-TKIs

Full size image

Table 3 shows the results of the analysis conducted using a Cox proportional hazards model conducted to identify the factors influencing the OS. The analysis identified the type of EGFR mutation in the tumor, the ECOG PS, the serum LDH level, treatment status for brain metastasis, and sequential therapy with osimertinib as factors that significantly influenced the OS. First-line treatment with afatinib or osimertinib showed a risk reduction versus treatment with a first-generation TKI, although the difference did not reach statistical significance.

Table 3 Associations of the patient characteristics with the OS (Cox proportional hazards model)
Full size table

Discussion

In the present study, we showed that sequential therapy with osimertinib in EGFR-mutant NSCLC patients with acquired T790M mutation was associated with a longer OS. The AURA study demonstrated that osimertinib yielded a longer PFS as compared to platinum-doublet therapy in patients with acquired resistance to primary EGFR-TKI therapy associated with emergence of the T790M mutation [7]. Therefore, the effectiveness of osimertinib for patients with acquired T790M mutation might be reflected in the results of the present study, with improvement of the prognosis of EGFR-mutant NSCLC with clinical application of sequential osimertinib therapy. However, it has also been reported that patients with acquired resistance to primary EGFR-TKI therapy associated with the emergence of the T790M mutation showed a longer PFS of first-line EGFR-TKI treatment [10,11,12,13] and a lower incidence of brain metastasis under the treatment with first- or second-generation EGFR-TKIs [14]. Therefore, in addition to the effectiveness of osimertinib, the selection of T790M-positive patients after first-line EGFR-TKI treatment might have resulted in the selection of patients with a favorable prognosis, resulting in a longer survival.

In the FLAURA trial, patients with untreated EGFR-mutant NSCLC who received first-line osimertinib treatment showed a significantly longer OS as compared to patients who received first-generation EGFR-TKIs as the first-line treatment [5]. In the present study, multivariate analysis revealed no significant association between the type of EGFR-TKI used for first-line treatment and the OS. This could be because some patients in whom the first-/second-generation EGFR-TKIs were changed to osimertinib due to the appearance of side effects rather than due to acquired resistance to the EGFR-TKIs have been included in the first- or second-generation EGFR-TKI group. In addition, the number of patients who were treated with osimertinib as first-line treatment was very small (n = 21), which could have led to a lack of statistical power.

Analysis of the OS data from the LUX-Lung 3, LUX-Lung 6 [15], and FLAURA [5] trials showed that the study treatment was more beneficial in patients with tumors carrying the exon 19 deletion as compared to exon 21 L858R as the EGFR mutation. The results of the present study showed that the hazard ratio for death was lower in patients with tumors carrying the exon 19 deletion. It is considered that the higher efficacy of EGFR-TKIs in patients with tumors carrying the exon 19 deletion might have contributed to the results of the present study.

Two previous studies have reported that combined therapy with erlotinib, a first-generation EGFR-TKI, plus a VEGF inhibitor yielded a longer PFS as compared to erlotinib monotherapy [16, 17]. In the present study, comparison by the log-rank test showed no association between combined first-generation EGFR-TKI plus VEGF inhibitor therapy and the OS. However, no definitive conclusions can be drawn because of the retrospective nature of the study and the small number of patients who received the combination therapy.

Numerous studies have reported serum LDH as a prognostic factor and indicated an association of the serum LDH with the OS in patients with NSCLC, whether treated by EGFR-TKI therapy [18], cytotoxic chemotherapy, or best supportive care [19]. As the mechanisms underlying this association, it is considered that tumor hypoxia is linked to glycolytic enzymes and angiogenetic factors, including LDH and VEGF, and that the invasive ability and angiogenic capability of tumors are promoted by acidification of the extracellular space induced by lactic acid [20, 21].

The development of brain metastasis has been reported to influence the clinical course of patients with EGFR-mutant NSCLC [14, 22]. In the present study, out of 37 patients with brain metastasis, 24 had received local therapy for brain metastasis. According to one previous report, the OS was longer in patients with brain metastasis who were treated with upfront radiotherapy as compared to systemic chemotherapy [23]. Because of the retrospective nature of our study, it would be difficult to exclude selection bias to definitively establish the superiority of the treatment option used in the present study. However, the present study findings do suggest that local therapy for brain metastasis might indeed prolong the survival in patients with EGFR-mutant NSCLC who develop brain metastasis.

There were several limitations of the present study. First, because of the retrospective nature of the study design, selection bias could have affected the results and we cannot conclude superiority of any treatment. Second, the survival rate might have been overestimated because of censoring of the patients when they were transferred to another hospital due to worsening of the general status. Third, none of the patients in this study had received combined therapy with gefitinib and cytotoxic agents, even though combined gefitinib plus chemotherapy has also been reported to yield a longer survival as compared to EGFR-TKI monotherapy [24]. Thus, we could not evaluate the association between this therapy and the patient survival.

In conclusion, the present study showed that patients developing acquired resistance to EGFR-TKI therapy associated with the emergence of the T790M mutation who received sequential therapy with osimertinib showed a longer OS and that EGFR mutation status, an ECOG PS, serum LDH level, and local therapy for brain metastasis were associated with OS. Our findings indicate that the development of new therapeutic strategies has led to the improved prognosis of patients with EGFR-mutant NSCLC in the clinical setting.