Progression of Central Nervous System Metastases in Advanced Nonsmall Cell Lung Cancer Patients Effectively Treated with First-Generation Epidermal Growth Factor Receptor-Tyrosine Kinase Inhibitor

Abstract

Central nervous system (CNS) progression is frequently detected in patients with favorable initial responses to first-generation epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs), but data are incomplete with respect to clinical features and prognostic factors of CNS failure in this population. We retrospectively evaluated 420 advanced nonsmall cell lung cancer (NSCLC) patients treated with first-generation EGFR-TKI for over 3 months. We analyzed CNS progression of these patients, defining as newly developed CNS metastases or progression of preexisting brain lesions after EGFR-TKI treatment. Of the 420 patients, 99 (23.6%) with CNS progression after EGFR-TKI initiation were identified. The median time to CNS progression was 12 months (95% confidence interval [CI] 10.5–13.5). Patients with L858R mutation were more likely to experience CNS failure than those with exon 19 deletion (p = 0.008). For patients with previous brain metastases (BM), the median time to CNS progression of patients with EGFR-TKI and local CNS therapy was significantly longer than those treated with EGFR-TKI alone (14.0 months vs. 11.2 months; p = 0.016). The median survival time after CNS progression was 11.2 months (95% CI 8.8–13.5). L858R mutation, multiple BM progression and CNS with other site failures were negative prognostic factor, while localized CNS therapy was the only significant favorable prognostic factor for overall survival after CNS progression. For advanced NSCLC patients treated successfully with first-generation EGFR-TKI, careful monitoring for CNS progression should be considered, especially for those with L858R mutation. Localized CNS therapy should be considered for CNS progression.

Introduction

Central nervous system (CNS) tumor metastases are frequent and serious complications of nonsmall cell lung cancer (NSCLC) that can decrease life quality and expectancy. Recently, with advances in neuroimaging techniques and prolonged survival resulting from new treatments, CNS metastases of NSCLC are increasingly diagnosed. Whole brain radiotherapy (WBRT), alone or combined with surgery and stereotactic radiosurgery (SRS), has been the standard treatment for patients with brain metastases (BM). However, the overall survival (OS) is only 4–5 months for unselected NSCLC patients treated with WBRT.¹

The first-generation tyrosine kinase inhibitors (TKIs) of the epidermal growth factor receptor (EGFR), such as gefitinb, erlotinib, and icotinib have offered therapeutic improvement for patients with NSCLC harboring EGFR mutations. Patients with NSCLC that harbor EGFR mutation can achieve better disease control and prolonged survival with EGFR-TKI. However, the prolonged survival of these patients with EGFR-TKIs may result in increased development of CNS metastases. EGFR-TKIs have distinct therapeutic benefits against BM for EGFR-mutant patients, offering an intracranial overall response rate of 85%, a disease control rate (DCR) of 94.6%, a median progression-free survival of 12.3 months, and a median survival time (mOS) of 16.2 months.² Thus, it may be reasonable to consider EGFR-TKIs for first-line treatment of patients with BM harboring EGFR mutations, which delay brain irradiation until CNS imaging or symptom progression, thereby avoiding WBRT-related neurocognitive side-effects.

CNS disease progression, either new lesions or the increase in preexisting brain lesions, frequently has been detected in patients with initially good responses to EGFR-TKIs.^3
–5 Data are incomplete regarding clinical features and prognostic factors of CNS failure for these patients, and patients with CNS metastases are under-represented in clinical trials of systemic administration. Thus, we performed a retrospective study to examine a large cohort of patients who had initially good responses to EGFR-TKI to assess the clinical features of CNS progression and clinical factors related to prognosis.

Materials and Methods

Study design and patients

A total of 420 NSCLC patients treated with first or second-line EGFR-TKI for over 3 months with good responses were reviewed at the Cancer center of West China Hospital, Sichuan University between June 2010 and June 2014. All patients received an initial gadolinium-enhanced brain magnetic resonance imaging (MRI) at the time of initial diagnosis of NSCLC. A subsequent brain MRI was performed if symptoms or signs were present to indicate CNS involvement. Medical records and imaging data for patients were collected to identify patients who developed CNS metastasis during their disease course, which was defined as the presence of one or more intra-axial enhancing lesions on MRI. Patients with leptomeningeal metastases (LM) were also included. CNS progression was defined as newly developed CNS metastases or progression of preexisting brain lesions after the response to first-generation EGFR-TKI.

We evaluated the treatment effects according to RECIST 1.1. This study was approved by the Institutional Review Board, and informed consent was obtained from each patient for the use of tissue in molecular analysis.

Mutation analysis

EGFR mutations were identified by real-time Amplification Refractory Mutation System (ARMS) quantitative polymerase chain reaction analysis using AmoyDxTM EGFR Mutation Detection Kit (Amoy Diagnostics Co., LTD, China) and EGFR Scorpion ARMS Kit (DxS Ltd., Manchester, United Kingdom).

Statistical analyses

The crude incidence of CNS progression was calculated as the proportion of patients with newly developed BM/LM or progression of preexisting CNS involvement to the number of patients treated with EGFR-TKI. Time to CNS progression was calculated from first-day of treatment with EGFR-TKI to the date of progression. OS after CNS progression was calculated from the time of CNS progression diagnosis until death from any cause. Patients alive at the time of statistical analysis were censored at last follow-up.

The Pearson chi-square test or the Fisher exact test was used to compare the frequency of CNS failure according to the risk factors. The Kaplan–Meier method was used to perform survival analysis, and survival curves were compared using the log-rank test. Multivariable predictors of survival were determined using Cox regression analysis. Two-sided values of p < 0.05 were considered statistically significant. All statistical analyses were performed with SPSS 20.0 software.

Results

Patient characteristics and clinical features of CNS progression after EGFR-TKI

Of the 420 patients, 99 who developed CNS metastases at any point in their disease course were identified, including 42 patients with parenchymal BM before EGFR-TKI treatment and 57 patients with no history of BM before the initiation of EGFR-TKI but who developed CNS metastasis after treatment. Therefore, 99 patients had CNS progression after initiation of treatment with EGFR-TKI and their characteristics are presented in Table 1. Isolated CNS failure, without systemic progression, was observed in 53 patients (53.5%). Most patients (79.8%) had numerous brain lesions. Patients (56/197) with L858R mutations were more likely to experience CNS failure, compared with 17.2% of patients (35/203) with exon 19 deletions (p = 0.008). The median follow-up time for all patients was 32.0 months (range, 6.5–101.0 months). The crude incidence of CNS progression in the cohort was 23.6%. The median time to CNS progression was 12 months (95% confidence interval [CI] 10.5–13.5). Of the 42 patients with parenchymal BM before EGFR-TKI treatment, 24 (57.1%) had received localized CNS therapy before or concomitant with EGFR-TKI, including 10 patients treated with SRS and 14 with WBRT. The median time to CNS progression of patients with local CNS therapy was significantly longer than those without local CNS therapy (14.0 months vs. 11.2 months; p = 0.016; Fig. 1).

FIG. 1.

For patients with previous brain metastases, the median time to CNS progression with local CNS therapy was significantly longer than those without local CNS therapy. CNS, central nervous system. Color image available online at www.liebertpub.com/cbr

Table 1.

The Characteristics of 99 Patients with Central Nervous System Progression After Initial Treatment with Epidermal Growth Factor Receptor-Tyrosine Kinase Inhibitor

Characteristic	CNS progression after initial EGFR-TKI therapy
Gender, n (%)
Male	31 (31.3)
Female	68 (68.7)
Smoking status, n (%)
Smoker	10 (10.1)
Nonsmoker	89 (89.9)
Pathological subtype, n (%)
Adenocarcinoma	96 (97)
Squamous cell carcinoma	3 (3)
Type of CNS progression, n (%)
Intracranial lesions	53 (53.5)
Intracranial and extracranial lesions	46 (46.5)
Symptoms of BM, n (%)
Yes	58 (58.6)
No	41 (41.4)
No. of BM, n (%)
1	20 (20.2)
>1	79 (79.8)
EGFR mutation type, n (%)
Exon 19 deletion	35 (35.4)
L858R mutation	56 (56.5)
Others	8 (8.1)
EGFR-TKI therapy
First line	54 (54.5)
Second line	45 (45.5)

CNS, central nervous system; EGFR-TKI, epidermal growth factor receptor-tyrosine kinase inhibitor.

Subsequent therapy after CNS progression in EGFR-TKI-treated patients

After CNS progression, 48 patients received localized CNS therapy. Among them, 36 patients received WBRT; 9 patients were treated with SRS; and 1 patient received SRS and then WBRT; 1 patient received SRS followed by BM resection; and 1 patient received only resection of a solitary BM.

Of the 53 patients with isolated CNS failure, 33 continued with gefitinib, 9 were treated with erlotinib, and 11 received chemotherapy. Of the 46 patients who had CNS and other site failures, 28 patients received chemotherapy and 4 patients were treated with erlotinib. The other 14 patients had no further systemic therapy.

Survival after CNS progression

At the final analysis (March 30, 2016), 26 patients were alive. The median survival time after CNS progression was 11.2 months (95% CI 8.8–13.5) (Fig. 2). In univariate analysis, the patients with exon 19 deletions had a longer OS than those with L858R mutations (22.8 vs. 9.8 months, p = 0.020, Fig. 3a). Patients with isolated CNS failure had greater OS than those with CNS and other site failures (13.6 vs. 8.5 months, p = 0.002, Fig. 3b). The OS of patients who had single brain metastasis was greater than that of patients with multiple BM (16.0 vs. 10.0 months, p = 0.011, Fig. 3c). Patients who received localized CNS therapy had increased survival, with a median survival of 13.6 months versus 8.0 months for patients without localized CNS therapy (p = 0.001, Fig. 3d). In multivariate analysis, the addition of localized CNS therapy was the only favorable independent predictor of better OS for these patients, with a hazards ratio (HR) of 0.33 (95% CI, 0.19–0.58, p = 0.002). Multiple BM (HR = 4.52; 95% CI, 1.35–15.16; p = 0.015) and L858R mutations (HR = 1.46; 95% CI, 1.01–2.12; p = 0.047) were identified as adverse prognostic factors for CNS progression (Table 2).

FIG. 2.

The median survival time after CNS progression. Color image available online at www.liebertpub.com/cbr

FIG. 3.

Survival factors after CNS progression by univariate analysis. (a) The patients with exon 19 deletions had a longer OS than those with L858R mutations. (b) Patients with isolated CNS failure had greater OS than those with CNS and other site failures. (c) The OS of patients who had single brain metastasis was greater than that of patients with multiple brain metastases. (d) Patients who received localized CNS therapy had increased survival compared with patients without localized CNS therapy. OS, overall survival. Color images available online at www.liebertpub.com/cbr

Table 2.

Univariate and Multivariate Analyses of Clinical Parameters on Overall Survival After Central Nervous System Progression (N = 99)

Factor	Univariate analysis			Multivariate analysis
Factor	HR	95% CI	p	HR	95% CI	p
Age (>65 vs. ≤65)	1.69	0.97–2.93	0.062
Smoker (yes vs. no)	1.15	0.55–2.43	0.705
EGFR mutations (L858R vs. 19-DEL)	1.52	1.06–2.18	0.024	1.46	1.01–2.12	0.047
Disease progression (isolated CNS vs. with other site)	2.41	1.31–3.52	0.003	1.60	0.83–3.09	0.159
BM No. (>1 vs. 1)	2.41	1.19–4.89	0.015	4.52	1.35–15.16	0.015
Local CNS therapy (yea vs. no)	0.31	0.18–0.54	0.011	0.33	0.19–0.58	0.002
Continued TKI after CNS failure (yea vs. no)	0.62	036–1.06	0.078

BM, brain metastase; CI, confidence interval; HR, hazard ratio.

Discussion

Here, we describe clinical features and prognostic factors of CNS progression in a cohort of patients who initially had good responses to first-generation EGFR-TKI but experienced CNS failure. To our knowledge, this is the largest study of this patient population and the first study to report predictors of survival after CNS progression in these patients. Our data will offer clinicians more information regarding CNS failure for patients who initially benefited from first-generation EGFR-TKI treatment.

Our analysis revealed that the crude incidence of CNS progression in patients who were effectively treated with first-generation EGFR-TKI was 23.6% after a median follow-up time of 32.0 months. Previous reports in patients who experienced CNS failure after clinical benefit from EGFR-TKIs have suggested a 26% to 28% crude incidence of overall CNS failure after a median follow-up of 21.4–42.2 months [4–5]. Our results therefore suggest a similar risk of CNS progression in this patient cohort. Approximately half of the patients had isolated CNS failure without systemic progression and most patients (79.8%) had numerous brain lesions. LM were observed in 29 patients as a relatively late event (median time 16.5 months) after EGFR-TKI treatment. Lee et al. reported that the median time to LM progression was 21.4 months and EGFR-TKIs did not affect the incidence or timing of LM development.⁶ Our study had too few patients with LM to confirm this (data not shown).

Our study showed that for patients with previous BM, the median time to CNS progression of patients with EGFR-TKI and local CNS therapy was significantly longer than those treated with EGFR-TKI alone (14.0 months vs. 11.2 months; p = 0.016). Zeng et al. also reported the median time to progression of BM was longer in gefitinib-WBRT group than gefitinib group.⁷ The attribution of the enhanced effectiveness might be synergistic effects of EGFR-TKIs and local CNS therapy. An interesting finding in our study was that patients with L858R mutations were more likely to experience CNS failure than those with exon 19 deletions (28.4% vs. 17.2%; p = 0.008). The patients with L858R mutations had a poor OS after CNS failure than those with exon 19 deletion (9.8 months vs. 22.8 months; p = 0.020). Several studies have indicated that advanced NSCLC patients with exon 19 deletion had a longer OS and progression-free survival (PFS) following treatment with gefitinib or erlotinib compared with those harboring L858R mutation.^8
–10 Up to present, there are no studies comparing the clinical differences between exon 19 deletion and L858R mutation in advanced NSCLC patients with CNS progression who had initially good responses to EGFR-TKI. We hypothesized that the lower risk of CNS progression in patients with exon 19 deletion after EGFR-TKI treatment contributed to PFS and OS advantage. Although we make note of this finding, given the small number of patients in our study and its retrospective nature, further prospective studies with greater numbers of patients with adequate follow-up are needed to define the predictive and prognostic roles of different EGFR mutations on the outcome of CNS progression after EGFR-TKIs treatment.

Understanding mechanism of CNS progression in patients effectively treated with EGFR-TKIs may help with treatment strategies. Two possible reasons for CNS progression may exist. First, EGFR-TKIs may incompletely penetrate the blood-brain barrier. McKillop et al. confirmed in a preclinical study that after oral administration of [¹⁴C]-gefitinib to albino and pigmented rats, radioactivity had limited distribution in the CNS.¹¹ Jackman et al. showed that greater than standard doses of gefitinib were insufficient for tumor growth inhibition in the cerebrospinal fluid (CSF) of one patient.¹² Some clinical studies suggest high dose pulsatile EGFR-TKIs are successful after CNS failure with standard dose EGFR-TKI.^13,14 However, inadequate drug delivery to the CSF might not be exclusively responsible for CNS failures with EGFR-TKI therapy. CNS metastases might acquire drug resistance during EGFR-TKI therapy, or they may be heterogeneous with the primary tumor in terms of their EGFR mutation status. Heon et al. reported the presence of a sensitizing exon 19 mutation concurrent with a T790M mutation in a brain specimen of a patient who experienced CNS failure with gefitinib therapy after initially responding.⁵ Ruppert et al. observed heterogeneity regarding EGFR mutation status of cranial and extracranial disease after treatment with gefitinib.¹⁵ Thus, both the unique microenvironment of the CNS and the intrinsic characteristics of metastatic CNS clones may contribute to different patterns of resistance. Studies are needed to characterize the molecular evolution of the primary tumor and CNS metastases in patients with EGFR mutations treated with EGFR-TKIs.

After CNS failure of standard EGFR-TKI therapy, different treatment patterns have been suggested. Some authors suggested using another TKI.^16
–18 Osimertinib is an oral EGFR-TKI, which is selective for T790M resistance mutations with activity in the pulmonary lesions and CNS metastases. Since osimertinib demonstrated greater penetration of blood-brain barrier than other EGFR-TKIs, and showed specific effects on T790M-positive lesions, osimertinib had significantly great efficacy in patients with advanced NSCLC with CNS metastases in whom disease had progressed during first-line EGFR-TKI therapy.^19,20

We also found that localized CNS therapy was the favorable independent predictor of OS for patients with CNS progression. Patients who received localized CNS therapy lived longer than those without localized CNS therapy. Recently, patients harboring EGFR mutation with cerebral metastases were recommended to receive EGFR-TKI first, in hopes of delaying or obviating the need for WBRT and the risks of neurocognitive side-effects from brain radiation.^21,22 Delaying WBRT for patients without obvious neurological signs or symptoms is an emerging concept. In our study, some patients treated with EGFR-TKIs eventually experienced CNS progression, and traditional methods such as WBRT or SRS were beneficial.

In vitro studies documented radiosensitivity of lung cancer cells with mutant EGFR.²³ Various reports indicate a favorable clinical effect with WBRT and EGFR-TKI therapy, which offers an excellent intracranial DCR.^24

–27 For EGFR-mutant patients with BM at initial staging, optimal management strategy with local radiation therapy and TKI should be evaluated in prospective clinical trials. The ongoing prospective TRACTS trial (clinicaltrials.gov/NCT01763385), to compare concurrent WBRT and erlotinib to erlotinib alone with WBRT at time of progression, may help us understand this issue.

We acknowledge that our study is limited by size (a single institution patient sample) and by its retrospective design. Our data should be confirmed with prospective studies.

Conclusions

The CNS was frequently a failure site for NSCLC patients treated otherwise successfully with EGFR-TKI. L858R mutations might be associated with higher risk of CNS progression and represent a negative prognostic factor, compared with exon 19 deletions. Careful monitoring for CNS failure and localized CNS therapy should be considered for CNS progression. Further prospective studies are needed to define the predictive and prognostic roles of different EGFR mutations on the outcome of CNS progression after EGFR-TKIs treatment.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Eichler

, Kahle

, Wang

, et al. EGFR mutation status and survival after diagnosis of brain metastasis in non-small cell lung cancer. Neuro Oncol, 2010; 12:1193.

Fan

, Xu

, Xie

. EGFR-TKI therapy for patients with brain metastases from non-small-cell lung cancer: A pooled analysis of published data. Onco Targets Ther, 2014; 10:2075.

Omuro

, Kris

, Miller

, et al. High incidence of disease recurrence in the brain and leptomeninges in patients with nonsmall cell lung carcinoma after response to gefitinib. Cancer, 2005; 103:2344.

Lee

, Choi

, Kim

, et al. Frequent central nervous system failure after clinical benefit with epidermal growth factor receptor tyrosine kinase inhibitors in Korean patients with non-small-cell lung cancer. Cancer, 2010; 116:1336.

Heon

, Yeap

, Britt

, et al. Development of central nervous system metastases in patients with advanced non-small cell lung cancer and somatic EGFR mutations treated with gefitinib or erlotinib. Clin Cancer Res, 2010; 16:5873.

Lee

, Han

, Kim

, et al. Impact of EGFR tyrosine kinase inhibitors versus chemotherapy on the development of leptomeningeal metastasis in never smokers with advanced adenocarcinoma of the lung. J Neurooncol, 2013; 115:95.

Zeng

, Zhang

, Liao

, et al. Gefitinib alone or with concomitant whole brain radiotherapy for patients with brain metastasis from non-small-cell lung cancer: A retrospective study. Asian Pac J Cancer Prev, 2012; 13:909.

Jackman

, Yeap

, Sequist

, et al. Exon 19 deletion mutations of epidermal growth factor receptor are associated with prolonged survival in non-small cell lung cancer patients treated with gefitinib or erlotinib. Clin Cancer Res, 2006; 12:3908.

Riely

, Pao

, Pham

, et al. Clinical course of patients with non-small cell lung cancer and epidermal growth factor receptor exon 19 and exon 21 mutations treated with gefitinib or erlotinib. Clin Cancer Res, 2006; 12:839.

10.

Fukuhara

, Maemondo

, Inoue

, et al. Factors associated with a poor response to gefitinib in the NEJ002 study: Smoking and the L858R mutation. Lung Cancer, 2015; 88:181.

11.

McKillop

, Hutchison

, Partridge

, et al. Metabolic disposition of gefitinib, an epidermal growth factor receptor tyrosine kinase inhibitor, in rat, dog and man. Xenobiotica, 2004; 34:917.

12.

Jackman

, Holmes

, Lindeman

, et al. Response and resistance in a non-small-cell lung cancer patient with an epidermal growth factor receptor mutation and leptomeningeal metastases treated with high-dose gefitinib. J Clin Oncol, 2006; 24:4517.

13.

Clarke

, Pao

, Wu

, et al. High dose weekly erlotinib achieves therapeutic concentrations in CSF and is effective in leptomeningeal metastases from epidermal growth factor receptor mutant lung cancer. J Neurooncol, 2010; 99:283.

14.

Grommes

, Oxnard

, Kris

, et al. Pulsatile high-dose weekly erlotinib for CNS metastases from EGFR mutant non-small cell lung cancer. Neuro Oncol, 2011; 13:1364.

15.

Ruppert

, Beau-Faller

, Neuville

, et al. EGFR-TKI and lung adenocarcinoma with CNS relapse: Interest of molecular follow-up. Eur Respir J, 2009; 33:436.

16.

Masuda

, Hattori

, Hamada

, et al. Erlotinib efficacy and cerebrospinal fluid concentration in patients with lung adenocarcinoma developing leptomeningeal metastases during gefitinib therapy. Cancer Chemother Pharmacol, 2011; 67:1465.

17.

Katayama

, Shimizu

, Suda

, et al. Efficacy of erlotinib for brain and leptomeningeal metastases in patients with lung adenocarcinoma who showed initial good response to gefitinib. J Thorac Oncol, 2009; 4:1415.

18.

Choong

, Dietrich

, Seiwert

, et al. Gefitinib response of erlotinib refractory lung cancer involving meninges-role of EGFR mutation. Nat Clin Pract Oncol, 2006; 3:50.

19.

Ballard

, Yates

, Yang

, et al. Preclinical comparison of osimertinib with other EGFR-TKIs in EGFR-mutant NSCLC brain metastases models, and early evidence of clinical brain metastases activity. Clin Cancer Res, 2016; 22:5130.

20.

Mok

, Wu

Y-L

, Ahn

M-J

, et al. Osimertinib or platinum-pemetrexed in EGFR T790M-positive lung cancer. N Engl J Med, 2017; 376:629.

21.

Kim

, Lee

, Choi

, et al. Epidermal growth factor receptor tyrosine kinase inhibitors as a first-line therapy for never-smokers with adenocarcinoma of the lung having asymptomatic synchronous brain metastasis. Lung Cancer, 2009; 65:351.

22.

Olson

, Paleologos

, Gaspar

, et al. The role of emerging and investigational therapies for metastatic brain tumors: A systematic review and evidence-based clinical practice guideline of selected topics. J Neurooncol, 2010; 96:115.

23.

Das

, Sato

, Story

, et al. Non-small-cell lung cancers with kinase domain mutations in the epidermal growth factor receptor are sensitive to ionizing radiation. Cancer Res, 2006; 66:9601.

24.

, Xu

, Deng

, et al. Treatment of brain metastasis from non-small cell lung cancer with whole brain radiotherapy and Gefitinib in a Chinese population. Lung Cancer, 2009; 65:198.

25.

Gow

, Chien

, Chang

, et al. Radiotherapy in lung adenocarcinoma with brain metastases: Effects of activating epidermal growth factor receptor mutations on clinical response. Clin Cancer Res, 2008; 14:162.

26.

Lind

, Lagerwaard

, Smit

, et al. Phase I study of concurrent whole brain radiotherapy and erlotinib for multiple brain metastases from non-small-cell lung cancer. Int J Radiat Oncol Biol Phys, 2009; 74:1391.

27.

Welsh

, Komaki

, Amini

, et al. Phase II trial of erlotinib plus concurrent whole-brain radiation therapy for patients with brain metastases from non-small-cell lung cancer. J Clin Oncol, 2013; 31:895.