Clinical management and outcome of patients with advanced NSCLC carrying EGFR mutations in Spain

Background Although the benefit of first-line epidermal growth factor receptor (EGFR) tyrosine-kinase inhibitors (TKIs) over chemotherapy has been demonstrated in several clinical trials, data from clinical practice is lacking and the optimal EGFR TKI to be used remains unclear. This study aims to assess the real-life diagnostic and clinical management and outcome of patients with advanced non-small-cell lung cancer (NSCLC) carrying EGFR mutations in Spain. Methods All consecutive patients recently diagnosed with advanced or metastatic NSCLC from April 2010 to December 2011 in 18 Spanish hospitals and carrying EGFR mutations were retrospectively evaluated. Results Between March and November 2013, a total of 187 patients were enrolled (98.3% Caucasian, 61.9% female, 54.9% never-smokers, 89.0% adenocarcinoma). Mutation testing was mainly performed on biopsy tumour tissue specimens (69.0%) using a qPCR-based test (90%) (47.0% Therascreen EGFR PCR Kit). Common sensitising mutations were detected in 79.8% of patients: 57.1% had exon 19 deletions and 22.6% exon 21 L858R point mutations. The vast majority of patients received first-line therapy (n = 168; 92.8%). EGFR TKIs were the most commonly used first-line treatment (81.5%), while chemotherapy was more frequently administered as a second- and third-line option (51.9% and 56.0%, respectively). Of 141 patients who experienced disease progression, 79 (56.0%) received second-line treatment. After disease progression on first-line TKIs (n = 112), 33.9% received chemotherapy, 8.9% chemotherapy and a TKI, and 9.8% continued TKI therapy. Most patients received first-line gefitinib (83.0%), while erlotinib was more frequently used in the second-line setting (83.0%). Progression-free survival (PFS) and overall survival (OS) in patients harbouring common mutations were 11.1 months and 20.1 months respectively (exon 19 deletions: 12.4 and 21.4 months; L858R: 8.3 and 14.5 months), and 3.9 months and 11.1 months respectively for those with rare mutations. Conclusion EGFR TKIs (gefitinib and erlotinib) are used as the preferred first-line treatment while chemotherapy is more frequently administered as a second- and third-line option in routine clinical practice in Spain. In addition, efficacy data obtained in the real-life setting seem to concur with data from EGFR TKI phase III pivotal studies in NSCLC. Electronic supplementary material The online version of this article (10.1186/s12885-018-4004-7) contains supplementary material, which is available to authorized users.


Background
Non-small-cell lung cancer (NSCLC) accounts for more than 85% of lung cancer cases [1] with the majority of patients presenting with advanced disease at the time of diagnosis [2]. Standard first-line treatment for advanced disease has usually consisted of conventional cytotoxic chemotherapy, mostly platinum-based regimens, although it provides limited benefits with regard to survival [3,4]. Advances in targeted and individualised treatment have led to the development of anti-epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs), such as first-generation TKIs (gefitinib, erlotinib) and second-line TKIs (afatinib), which irreversibly bind to the tyrosine kinase receptor. In addition, said advances have brought about the recently-available thirdgeneration TKIs such as osimertinib, an oral, irreversible EGFR-TKI that is selective for both EGFR and T790 M resistance mutations and acts on the central nervous system. The development of these treatment strategies has markedly improved both clinical management and the outcome of patients with advanced NSCLC.
A large body of randomised clinical trials demonstrated the superior clinical effectiveness of EGFR TKIs compared with standard chemotherapy in patients with stage IIIB or IV NSCLC whose tumours tested positive for EGFR mutations [7-10, 13, 16, 17]. A large proportion of studies with these targeted agents were carried out in Asian patients from Japan, China and South Korea, where the incidence of EGFR mutations was high. However, some studies demonstrating the clinical benefit of TKIs over chemotherapy have been conducted in Caucasian populations (EURTAC [7], IFUM [18]), therefore confirming that the presence of EGFR mutations and not ethnicity is the most reliable factor predicting sensitivity to EGFR TKIs. All studies reported a superior benefit in overall response rate (ORR) and improvement [7-10, 13, 16, 17] in progression-free survival (PFS) for patients with EGFR mutation-positive NSCLC treated with TKIs compared with standard chemotherapy. None of the individual studies found a significant difference in overall survival (OS) between TKIs and chemotherapy, probably due to subsequent treatments and the high degree of crossover that may have confounded the effect of the initial first-line treatment.
The enhanced response to EGFR TKIs in patients harbouring activating mutations led to the recommendation of upfront EGFR mutation testing to guide therapeutic decision-making [19][20][21][22]. However, there is no consensus on the optimal detection method to identify EGFR mutations [23,24] and the sources of tumour material (biopsy tumour tissue samples, cytology specimens or serum samples) have been a notable consideration in EGFR mutation detection.
Although the benefit of EGFR TKIs over chemotherapy has been clearly demonstrated in the first-line setting in several clinical trials, data from clinical practice is lacking and there are still some concerns regarding the optimal EGFR TKI to be used. Moreover, the most beneficial therapy (EGFR TKIs or chemotherapy) and the role of EGFR mutation status in second-line treatment and beyond still remain the subject of debate.
In addition to the lack of data available from routine clinical practice in patients with EGFR-mutated NSCLC, particularly in Caucasians, there is an unmet need for real-life data on treatment patterns and outcome at a national level in Spain. This study was conducted to assess the clinical management and outcome of patients with advanced NSCLC carrying EGFR-positive mutations in the real-world clinical setting in Spain.

Study design and patients
This was a multicentre, retrospective, observational study conducted in 18 hospitals throughout Spain. All adult patients (aged ≥18 years) recently diagnosed with histologically or cytologically confirmed advanced NSCLC from April 2010 to December 2011 and carrying EGFR-positive mutations were retrospectively evaluated.
The study was carried out in accordance with the Declaration of Helsinki and applicable regulatory requirements. Approval of the study protocol was obtained from the Hospital del Mar Clinical Research Ethics Committee (Barcelona, Spain). Written informed consent was obtained from all patients to retrospectively collect data from medical charts.
The primary endpoint of the study was to describe the diagnostic and clinical management patterns of patients with advanced or metastatic NSCLC carrying EGFR-positive mutations. For this purpose, EGFR mutation testing methods, source of tumour material, treatment setting and therapeutic strategies were analysed. Secondary endpoints included the clinical outcome (ORR, PFS and OS) of the overall population according to the line of therapy and treatment received (chemotherapy or EGFR TKIs [gefitinib or erlotinib]), the type of EGFR mutation (common or rare sensitising mutations, exon 19 deletion or L858R point mutation), and other relevant clinical characteristics (i.e. Eastern Cooperative Oncology Group [ECOG] performance status).

Statistical analysis
A descriptive analysis was performed of diagnostic and clinical management variables collected from patient medical records. Quantitative variables were described using measures of central tendency and dispersion (mean, median, standard deviation [SD], minimum, maximum, first quartile and third quartile) and the results are expressed as mean ± SD or median (range). Qualitative variables are presented as absolute and relative frequencies. Efficacy analyses were conducted on the patients who had available data from at least one evaluation of response (8 weeks for EGFR TKI and 6 weeks for chemotherapy). Tumour response was assessed based on the unidimensional Response Evaluation Criteria In Solid Tumors (RECIST) version 1.1 [25] if the disease was measurable or by the investigator in those patients with a non-measurable disease according to local practice. ORR was calculated as the sum of patients achieving complete response and partial response as the best response achieved. PFS was assessed from the start of therapy for NSCLC until documented disease progression or death from any cause. Patients were censored at the date of last follow-up if still alive or without disease progression at the time of the analysis. OS was calculated as the time elapsed from the start of treatment to death. Patients were censored at the date of last follow-up if still alive at the time of the analysis. The probability of PFS and OS was estimated using the Kaplan-Meier method.
The statistical analysis was performed using the statistical package SAS version 9.02.

Patient population
A total of 187 newly diagnosed advanced NSCLC patients (from April 2010 to December 2011) from 18 Spanish sites were retrospectively evaluated between March and November 2013. Six patients were excluded as they did not meet the inclusion criteria (the informed consent of two patients was not available and four patients were not diagnosed between April 2010 and December 2011). The demographic and clinical characteristics of the 181 evaluable patients are shown in Table 1. Briefly, 61.9% were female, 98.3% were Caucasian and 54.9% were neversmokers. The most frequent comorbidities were hypertension (48.1%), dyslipidaemia (21.0%), diabetes (14.4%), cardiovascular disease (11.6%), and chronic obstructive pulmonary disease (COPD) (10.5%). The most common histological type was adenocarcinoma (89.0%). ECOG performance status at diagnosis of advanced disease was 0 or 1 in 80.1% of patients. Most patients (87.8%) had stage IV disease at diagnosis. Metastases were mainly located in the lungs (45.7%), bone (42.9%) and pleura (28.6%).

EGFR mutation analysis
Mutation testing was mainly conducted in external laboratories (68.0%). The median time elapsed from the date the sample was sent to the laboratory until the results were obtained (turnaround time [TAT]) was 8.5  Table 3.

Treatment
The vast majority of the patients had received first-line treatment after diagnosis of advanced NSCLC (92.8%). TKIs were used as first-line treatment in the majority of patients (81.5%), while chemotherapy-based regimens were more commonly administered as second-and third-line options (51.9% and 56.0%, respectively). Firstline chemotherapy followed by maintenance EGFR TKIs was used in less than 5% of patients. Of the 168 patients who received first-line treatment, 79 (47.0%) underwent second-line treatment and 25 (14.9%) and 10 (6.0%) received third-and fourth-line treatment, respectively (Table 4).
Of 141 patients who experienced disease progression on first-line treatment, 79 (56.0%) patients received second-line treatment. After disease progression on firstline EGFR TKIs (n = 112), 33.9% received chemotherapy, 8.9% chemotherapy and TKI, and 9.8% received further treatment with single agent TKI therapy. The majority of patients received first-line gefitinib treatment (83.2%), while erlotinib was the most frequent TKI used in the second-line setting (83.3%).
Of patients who received first-line chemotherapy (18.5%), doublet chemotherapy was used in 77.4% and 60.5% of patients as first and second-line treatment option, respectively.
EGFR TKI treatment resulted in an ORR of 53.1% in patients with common sensitising EGFR mutations who received TKIs while the ORR for those receiving chemotherapy was 18.8% ( Table 6). The median PFS in patients carrying common sensitising mutations and treated with first-line TKIs was 11.1 (95% CI: 9.3-12.7) months while those receiving chemotherapy showed a PFS of 5.8 (95% CI: 4.2-7.6) months. The median OS was 20.1 (95% CI: 15.7-31.5) months with first-line EGFR TKIs and 12.1 (95% CI: 9.4-not achieved) months with chemotherapy.

Discussion
The present study examined the patterns of diagnostic and clinical management of patients with NSCLC carrying EGFR-positive mutations in routine clinical practice in Spain. As expected for patients carrying EGFR mutations and in line with previous reports in Caucasians [18], we found a population with a high proportion of females, never-smoker patients, and adenocarcinoma histology. Additionally, consistent with previous data in Caucasian patients, the majority of EGFR activating mutations were exon 19 deletions (57.1%) and exon 21 L858R point mutations (22.6%) [12]. As seen in our study, a variety of methodologies are used for EGFR mutation detection, as there is currently no consensus on the optimal method or source used for testing. The heterogeneity in the detection method has a potential relevance due to the differences in sensitivity between methods. Direct sequencing of DNA has traditionally been the "gold standard" for EGFR mutation testing, though this method is mainly limited by its   [26,27]. Indeed, only a small proportion of tumour samples (11.0%) were analysed using direct sequencing in daily-care patients analysed. EGFR mutation testing was primarily performed using RT-PCR-based tests (80.0%), the Therascreen ® EGFR Mutation Test kit (ARMS) being the most commonly applied mutation testing assay, which has demonstrated improved sensitivity and TAT [28]. Paraffin-embedded tumour tissue specimens have conventionally been the main source of tumour material for EGFR mutation testing in lung cancer and they currently still account for the majority of diagnostic samples in the clinical practice. Cytology specimens have been proven to be an adequate alternative source for mutation testing when tissue samples are not available or have a low content of tumour DNA [29,30] and their use has increased over recent years. In our study, tumour tissue from bronchial biopsy was the most frequently used source of tumour material for EGFR mutation analysis (70.0%) while cytology specimens were used in about one third of patients at the time when the study was carried out. The median TAT of 9 days seen in our study for sample analysis performed externally demonstrates a well-structured set up for this analysis in Spain. This allows physicians to have information available for treatment decision-making within an adequate time period after diagnosis, even in centres without local facilities to perform the analysis.
With regard to clinical management, one of the main findings of our study is the high proportion of patients who received EGFR TKIs in the first-line setting (82.0%), even though some of the current targeted agents were not available for patients harbouring EGFR mutations (e.g. erlotinib and afatinib) or had recently been approved at the time when the patients were diagnosed (e.g., gefitinib was marketed in March 2010). As expected, the most frequently used first-line TKI was gefitinib (83.2%). Therefore, EGFR TKI treatment may have been introduced early in the therapeutic armamentarium for advanced NSCLC in Spain. These findings therefore suggest that EGFR mutation testing may have been adopted early as a routine procedure to guide therapeutic decision-making in clinical practice in Spain even before it was widely adopted and recommended by major oncology groups, including the Spanish Society of Medical Oncology (SEOM) and the Spanish Society of Pathology (SEAP) [19][20][21]31]. While TKIs were used as first-line treatment in the majority of patients, chemotherapy-based regimens were the preferred second-line option in our series. Only a small proportion of patients continued EGFR TKI therapy after disease progression, with a similar number of patients receiving single-agent TKI and TKIs plus chemotherapy. Acquired resistance to EGFR TKIs is the main limitation to a long-lasting benefit of these targeted agents in patients with EGFR mutation-positive NSCLC [32], with the EGFR T790 M mutation being responsible for resistance in up to 60% of cases [33]. However, withdrawal of an EGFR TKI at the onset of resistance may lead to rapid tumour growth [34,35]. The potential benefit of continuing EGFR TKI treatment beyond disease progression has been addressed in several studies in the last years [36][37][38]. The single-arm phase II study ASPIRATION supports the feasibility of continuation of single-agent erlotinib beyond disease progression in patients with EGFR mutation-positive NSCLC [37]. However, further research based on randomised studies is needed before firm conclusions can be drawn. The phase III IMPRESS study showed that continuation of gefitinib in combination with platinum-based doublet chemotherapy after disease progression on first-line gefitinib did not prolong PFS compared with chemotherapy alone in patients with NSCLC carrying EGFR mutations [38]. The third-generation EGFR inhibitors, which can selectively target both sensitising mutations and the T790 M mutation, have demonstrated the benefit of continuing EGFR TKI treatment beyond progression for patients with T790 M mutation-positive NSCLC [39]. The recent approval of the third-generation EGFR TKI osimertinib may change the treatment paradigm after disease progression on EGFR TKI treatment in patients with T790 M mutation-positive NSCLC for which no other resistance mechanisms are identified.
Regarding the efficacy data, we found that tumour response and survival seem to be similar between gefitinib and erlotinib in real-life patients. However, the differences in the proportion of patients receiving first-line gefitinib and erlotinib in our series and the lack of matched comparisons make it difficult to obtain reliable data from our descriptive analysis. As a descriptive comparison only, considering the obvious limitations, the PFS achieved in this study with gefitinib (9.9 months) was within the range reported in the clinical trials carried out with gefitinib in Asian patients with advanced NSCLC [16,40,41] and the PFS data reported with gefitinib in a European population of Caucasian patients with advanced NSCLC harbouring EGFR mutations [18]. Similarly, erlotinib resulted in a comparable PFS to that reported in clinical trials with this targeted agent in Caucasians [7].
As expected, efficacy figures seem to be superior for patients harbouring common sensitising mutations in relation to those with rare mutations. Of note, only 12.5% of the patients carrying rare sensitising mutations (G719X) responded to EGFR-TKIs. Such a low response rate raises the question of whether these mutations should be considered "sensitizing" at all when it comes to EGFR first-generation TKIs.
Furthermore, PFS and OS appear to be longer for EGFR TKI-treated patients carrying exon 19 deletions compared with those with L858R point mutations. These findings are in line with previous clinical trials where numerical but non-significant differences in PFS were shown between patients treated with gefitinib harbouring exon 19 deletions and those with the L858R point mutation (11.5 months vs. 10.8 months, p = 0.90 in the NEJ002; Hazard ratio [HR] = 1.13, 95% CI = 0.63-2.03, p = 0.68 in the WJTOG3405) [5,7,16]. Similarly, a beneficial effect in favour of patients receiving the recently introduced TKI afatinib and carrying the exon 19 deletions was reported [8,9].
In addition to the obvious limitations, arising from the retrospective nature of the study, the authors acknowledge that one of the main limitations of this study is the incorporation of the second-generation EGFR TKI afatinib for EGFR-mutated NSCLC in the last years which may have changed the prescription patterns regarding the type of EGFR TKI used. Despite these limitations, our findings still offer a valid global picture regarding the management of EGFR-mutant NSCLC patients who typically receive EGFR TKIs as their initial therapy. In addition, this study might offer a welcome addition to the limited "real-world" data on treatment patterns and clinical outcome of patients carrying EGFR-positive mutations in clinical practice, particularly in Caucasians. Our national data collection therefore provides an interesting overview of real-life clinical practice for the management of EGFR-mutated NSCLC in Spain.

Conclusions
To our knowledge, this is the first study to have focused on the clinical management and outcome of real-life patients with advanced EGFR-mutated NSCLC in Spain. Our data show that EGFR TKIs were used as the preferred first-line treatment while chemotherapy was more frequently administered as a second-and third-line option. In addition, efficacy data, in terms of PFS and OS, obtained from our national real-world data collection, seem consistent with data from EGFR TKI phase III pivotal studies in NSCLC.

Funding
The study was supported by AstraZeneca. This company has participated in the design of the study, data analysis and interpretation of the data, and in the preparation of the manuscript.

Availability of data and materials
The dataset supporting the conclusions of this article is included as an Additional file 1. have made substantial contributions to the study design and to the analysis and interpretation of the data. They have critically reviewed the manuscript and given approval to the submitted and final versions. They have also contributed to the inclusion of patients and acquisition of data. MLA, has made a substantial contribution to the analysis and interpretation of the data, and has critically reviewed the manuscript and given approval to the submitted and final versions.

Ethics approval and consent to participate
The study was approved by the Hospital del Mar Clinical Research Ethics Committee (Barcelona, Spain). Written informed consent was obtained from all the participant patients.

Consent for publication
Not applicable because this manuscript does not contain any individual person's data.
Competing interests M.L. Amador is an employee of the study sponsor. The remaining authors declare that they have no competing interests.

Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.