Afatinib 30 mg in the treatment of common and uncommon EGFR-mutated advanced lung adenocarcinomas: a retrospective, single-center, longitudinal study.

Background: Afatinib 30 mg has been proved to be with comparable efficacy but more tolerable than the dose of 40 mg for Asian patients with non-small cell lung cancer (NSCLC). This study aimed to investigate the clinical outcomes of afatinib at 30 mg/d in the treatment of advanced lung adenocarcinomas (LAD) with common and uncommon epidermal growth factor receptor (EGFR) mutations.
Methods: EGFR-mutated advanced LAD patients receiving afatinib (30 mg/d) from January 2017 to November 2021 were retrospectively included. EGFR status was classified into three subtypes, namely common mutations including exon 19 deletions (19del) and exon 21 L858R (21L858R), uncommon mutations including G719X, L861Q, S768I, and complex mutations, and separately exon 20 insertions (20ins). Progression-free survival (PFS), objective response rate (ORR), disease control rate (DCR) and adverse events (AEs) were analyzed during regular follow-up.
Results: The overall median PFS of totally 58 included patients was 9.83 [95% confidence index (CI): 5.76-13.91] months. The number of patients with common, uncommon, and 20ins mutations was 32 (55.2%), 19 (32.8%) and 7 (12.1%), respectively. Baseline characteristics did not differ significantly among the three subtypes. The corresponding median PFS was 13.97 (12.06-15.89), 8.48 (0.32-16.64), and 3.78 (1.93-5.64) months, respectively (P=0.002). In the first-line setting, patients with common and uncommon mutations had a significantly longer PFS compared to those with 20ins [14.53 (13.53-15.53) vs. 10.39 (4.87-15.91) vs. 2.37 (0.00-5.11) months, P<0.001]. The first-line ORR showed significant differences among the three subtypes (60% vs. 80% vs. 0.0%, P=0.023). All-grade AEs occurred in 22 patients (37.9%). AEs ≥ grade 3 mainly included diarrhea (8.6%), and none of the patients discontinued treatment due to severe AEs. Conclusions: Afatinib at 30 mg/d is associated with a favorable efficacy and tolerability in the treatment of advanced LAD with common and major uncommon EGFR mutations except 20ins. Further large-scale prospective studies are warranted to confirm our findings. 2022 Journal of Thoracic Disease. All rights reserved.

Introduction

Epidermal growth factor receptor-tyrosine kinase inhibitor (EGFR-TKI) is the standard treatment for lung adenocarcinomas (LAD) with EGFR mutations (1). While all generations of EGFR-TKIs are applicable for EGFR-mutated LAD, the effectiveness may vary across different EGFR mutation subtypes. EGFR-TKI-sensitive mutations exon 19 deletions (19del) and exon 21 L858R (21L858R) are the most common mutations in non-small cell lung cancer (NSCLC) (2,3). However, uncommon EGFR are increasingly reported owing to the technical advances in genomic sequencing (4,5). These uncommon mutations primarily consisting of G719X in exon 18, S768I in exon 20, L861Q in exon 21, classical compound mutations, and exon 20 insertions (20ins) account for 10–20% of all EGFR mutations. Inconsistent efficacy to EGFR-TKIs has been observed in LAD patients with uncommon EGFR mutations and treatment strategies for this entity remain to be optimized (4).

Increasing evidence has shown favorable outcomes of LAD with uncommon EGFR mutations upon treatment with second-generation TKI afatinib which has an irreversible and broad inhibitory spectrum against the ErBB family (6-8). Afatinib gained similar or even better effectiveness to first-generation EGFR-TKIs in the treatment of NSCLC with EGFR exon 19del or 21L858R mutations (9,10). Moreover, afatinib is superior to the first-generation TKIs gefitinib and erlotinib for major uncommon mutations including G719X, S768I, L861Q, and classical compound mutations (7,11). A combined post-hoc analysis of LUX-Lung series trials demonstrated that afatinib treated NSCLC patients with G719X, S768I and L861Q had an objective response rate (ORR) of 77.8%, 100% and 56.3%, and a PFS of 13.8, 14.7 and 8.2 months, respectively (6). However, clinical benefit of EGFR-TKIs seemed to be inferior in patients with EGFR 20ins mutation (6).

The clinical application of afatinib may be limited due to its high frequency of adverse events (AEs), especially for Asian patients. The recommended initiating dose of afatinib at 40 mg/d was associated with up to 28% of grade 3 or greater treatment-related AEs in the LUX-Lung series studies (3,10,12-14). Asian patients seemed to be more susceptible to experiencing AEs caused by 40 mg afatinib compared with non-Asian patients (13,15). In the real-world studies, 29.6–38.3% of Asian patients receiving afatinib starting from 40 mg needed a dose reduction throughout the course of treatment (16,17). Recent studies demonstrated that afatinib at 30 mg/d was more tolerable than the dose of 40 mg/d, but favorable efficacy was maintained for Asian patients with NSCLC (18-20). However, evidence is sparse regarding the efficacy and safety of afatinib at 30 mg in the treatment of advanced LAD with uncommon EGFR mutations.

Hence, in the present study on EGFR-mutated LAD, we attempted to investigate the clinical outcomes of afatinib starting at 30 mg/d on different EGFR subtypes. We hypothesized that afatinib 30 mg would be effective and well-tolerable for either common or major uncommon EGFR mutations. We present the following article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-22-507/rc).

Methods

Patients and treatment

This was a single-center, retrospective, longitudinal study. Medical records of advanced NSCLC patients with EGFR mutation who received afatinib at a starting dose of 30 mg from January 2017 to November 2021 at Shanghai Chest Hospital, China were retrospectively reviewed. The inclusion criteria were: (I) pathologically confirmed stage IIIB/IIIC/IV LAD according to the 8th edition of the American Joint Committee on Cancer (AJCC) staging manual; (II) received afatinib as a first- or later-line therapy for at least 30 consecutive days; (III) with EGFR mutations including 19del, 21L858R, G719X, S768I, L861Q, classical compound mutations, and exon 20ins mutation. Patients without complete medical records or follow-up information were excluded.

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The institutional review board of Shanghai Chest Hospital approved the study (No. IS22027), and all the patients supplied written consent before treatment.

Data collection and follow-up

Baseline clinicopathological data including age, gender, smoking history, Eastern Cooperative Oncology Group performance status (ECOG PS) score, clinical stage, treatment line, and metastatic site at initiation were collected from the medical record system. EGFR genotyping was performed at the hospital’s laboratory by polymerase chain reaction (PCR) or next-generation sequencing (NGS) as previously described (21). EGFR mutations were divided into the following three subtypes: (I) common mutations, including exon 19del and 21L858R; (II) uncommon mutations, including G719X in exon 18, L861Q in exon 21, S768I in exon 20, and classical complex mutations; and (III) exon 20ins.

Therapeutic and prognostic information was retrospectively collected. Efficacy was assessed according to the Response Evaluation Criteria in Solid Tumor (version 1.1) with complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD) by computed tomography (CT) scans, nuclear magnetic resonance imaging (MRI) or abdominal ultrasound at every 4–8 weeks during treatment (22). ORR was defined as achieving CR or PR. Progression-free survival (PFS) was defined as the time from starting administration of oral afatinib to PD or last follow-up. AEs were graded using the Common Terminology Criteria for Adverse Events (CTCAE), version 4.0.

Statistical analysis

Continuous variables were expressed as mean ± standard deviation or median and interquartile range (IQR). Categorical variables were reported as number and percentage. The chi-square test was used to compare the subgroup differences in baseline characteristics. PFS was analyzed by the Kaplan-Meier survival curves and subgroup comparsion was conducted using a log-rank test. Multivariable Cox regression analysis was used to determine the estimated hazard ratio (HR) for PFS. Analyses were performed using SPSS 23.0 software. A two-sided P value <0.05 was considered statistically significant.

Results

Baseline characteristics

Among 62 initially screened patients, four patients were excluded due to incomplete medical records or follow-up information. A total of 58 patients were finally included in this study. The median age was 62 (range, 41–78) years old. Male patients accounted for 53.4%, and 26 (44.8%) patients were ever or current smokers. Half of all patients received afatinib as the first-line regimen. Eleven (19.0%) patients had brain metastases. Most patients (98.3%) had an ECOG PS score of 0–1 except 1 patient had a score of 2. At the data cut-off date, the median follow-up duration was 22.9 months ().

Table 1

Comparison of baseline characteristics among patients with common and uncommon EGFR mutations

Characteristics	19del/21L858R (n=32), n (%)	G719X/L861Q/S768I/complex (n=19), n (%)	20ins (n=7), n (%)	P value
Age (years)				0.601
<65	20 (62.5)	12 (63.2)	3 (42.9)
≥65	12 (37.5)	7 (36.8)	4 (57.1)
Sex				0.446
Male	15 (46.9)	11 (57.9)	5 (71.4)
Female	17 (53.1)	8 (42.1)	2 (28.6)
Smoking history				0.451
No	20 (62.5)	9 (47.4)	3 (42.9)
Ever/current	12 (27.5)	10 (52.6)	4 (57.1)
Stage				0.155
IIIB/IIIC	2 (6.3)	4 (21.1)	0 (0.0)
IV	30 (93.8)	15 (78.9)	7 (100.0)
Treatment line				0.852
First-line	15 (46.9)	10 (52.6)	4 (57.1)
≥ Second-line	17 (53.1)	9 (47.4)	3 (42.9)
Brain metastasis				0.376
No	28 (87.5)	14 (73.7)	5 (71.4)
Yes	4 (12.5)	5 (26.3)	2 (28.6)
Number of metastatic sites				0.702
<3	27 (84.4)	15 (78.9)	5 (71.4)
≥3	5 (15.6)	4 (21.1)	2 (28.6)

EGFR, epidermal growth factor receptor; 19del, exon 19 deletions; 21L858R, exon 21 L858R; 20ins, exon 20 insertions.

The number of patients with EGFR common mutations of 19del or 21L858R was 32 (55.2%). The number of patients with EGFR uncommon mutations of G719X, L861Q, S768I, or complex mutations was 19 (32.8%), while the number of patients with EGFR 20ins was 7 (12.1%). The baseline characteristics among the three subtypes of EGFR mutations did not differ significantly ().

Efficacy evaluation

The overall median PFS was 9.83 [95% confidence index (CI): 5.76–13.91] months. A total of 46 (79.3%) patients experienced disease progression or died at the last follow-up and 12 (20.7%) patients were still under treatment.

Comparisons between subgroups showed that significant differences in PFS and ORR existed among patients with different subtypes of EGFR mutations. The median PFS for patients with EGFR common mutations (19del/21L858R), uncommon mutations (G719X/L861Q/S768I, or complex mutations), and EGFR 20ins was 13.97 (12.06–15.89), 8.48 (0.32–16.64), and 3.78 (1.93–5.64) months, respectively (P=0.002) (). The ORR for subgroups of common mutations, uncommon mutations, and EGFR 20ins was 43.8%, 63.2%, and 0.0%, respectively (P=0.016). The corresponding DCR was 84.4%, 94.7%, and 85.7%, respectively (P=0.537) ().

Figure 1

Comparison of survival curves among patients with common and uncommon EGFR mutations in the overall population (A) and the first-line (B). PFS, progression-free survival; CI, confidence index; m, months; 19del, exon 19 deletions; 21L858R, exon 21 L858R; 20ins, exon 20 insertions; HR, hazard ratio; EGFR, epidermal growth factor receptor.

Table 2

Efficacy of afatinib-treated EGFR-mutated advanced LAD

Characteristics	N	PFS			ORR			DCR
		PFS (95% CI) (months)	P value		N (%)	P value		N (%)	P value
Age (years)			0.140			0.710			0.692
<65	35	9.44 (3.53–15.35)			15 (42.9)			30 (85.7)
≥65	23	13.25 (6.04–20.46)			11 (47.8)			21 (91.3)
Sex			0.114			0.125			1.000
Male	31	8.48 (3.56–13.41)			11 (35.5)			27 (87.1)
Female	27	11.38 (6.93–15.82)			15 (55.6)			24 (88.9)
Smoking history			0.194			0.380			1.000
No	32	11.38 (4.91–17.84)			16 (50.0)			28 (87.5)
Ever/current	26	8.48 (5.00–11.96)			10 (38.5)			23 (88.5)
Stage			0.425			0.393			1.000
IIIB/IIIC	6	12.13 (1.93–22.33)			4 (66.7)			6 (100.0)
IV	52	9.44 (5.81–13.06)			22 (42.3)			45 (86.5)
EGFR mutation			0.002			0.016			0.537
19del/21L858R	32	13.97 (12.06–15.89)			14 (43.8)			27 (84.4)
G719X/L861Q/S768I/complex mutations	19	8.48 (0.32–16.64)			12 (63.2)			18 (94.7)
20ins	7	3.78 (1.93–5.64)			0 (0.0)			6 (85.7)
Treatment line			0.464			0.035			0.102
First-line	29	11.38 (7.96–14.79)			17 (58.6)			28 (96.6)
≥ Second-line	29	6.67 (0.67–12.68)			9 (31.0)			23 (79.3)
Brain metastasis			0.097			1.000			0.327
No	47	10.39 (6.68–14.10)			21 (44.7)			40 (85.1)
Yes	11	5.33 (1.74–8.91)			5 (45.5)			11 (100.0)
Number of metastatic sites			0.030			0.517			0.607
<3	47	11.38 (4.94–17.81)			20 (42.6)			42 (89.4)
≥3	11	5.52 (0.00–11.62)			6 (54.5)			9 (81.8)

EGFR, epidermal growth factor receptor; LAD, lung adenocarcinomas; PFS, progression-free survival; ORR, objective response rate; DCR, disease control rate; CI, confidence index; 19del, exon 19 deletions; 21L858R, exon 21 L858R; 20ins, 20 exon insertions.

For the 29 patients who received first-line afatinib, the median PFS was 11.38 (7.96–14.79) months. Patients with EGFR 20ins had a significantly poorer PFS compared to patients with other EGFR common and uncommon mutations (2.37 vs. 14.53 vs. 10.39 months, P<0.001) (). Similarly, patients with EGFR 20ins had a remarkably lower ORR (0% vs. 60% vs. 80%, P=0.023) and DCR (75% vs. 100% vs. 100%, P=0.039) compared to patients with common and uncommon EGFR mutations ().

Table 3

Efficacy of afatinib 30 mg treatment in the first-line

Characteristics	N	PFS			ORR			DCR
		PFS (95% CI) (months)	P value		N (%)	P value		N (%)	P value
Age (years)			0.707			0.876			0.483
<65	15	11.18 (2.84–19.52)			9 (60.0)			15 (100.0)
≥65	14	13.25 (6.19–20.31)			8 (57.1)			13 (92.9)
Sex			0.212			0.176			1.000
Male	15	11.18 (5.85–16.51)			7 (46.7)			14 (93.3)
Female	14	12.13 (8.96–15.30)			10 (71.4)			14 (100.0)
Smoking history			0.233			0.219			0.448
No	16	12.13 (8.96–15.30)			11 (68.8)			16 (100.0)
Ever/current	13	11.18 (6.02–16.34)			6 (46.2)			12 (92.3)
Stage			0.782			1.000			1.000
IIIB/IIIC	5	12.13 (0.00–33.31)			3 (60.0)			5 (100.0)
IV	24	11.38 (7.34–15.41)			14 (58.3)			23 (95.8)
EGFR mutation			<0.001			0.023			0.039
19del/21L858R	15	14.53 (13.53–15.53)			9 (60.0)			15 (100.0)
G719X/L861Q/S768I/complex mutations	10	10.39 (4.87–15.91)			8 (80.0)			10 (100.0)
20ins	4	2.37 (0.00–5.11)			0 (0.0)			3 (75.0)
Brain metastasis			0.043			0.683			1.000
No	21	13.25 (9.00–17.50)			13 (61.9)			20 (95.2)
Yes	8	5.33 (0.86–9.79)			4 (50.0)			8 (100.0)
Number of metastatic sites			0.238			0.622			1.000
<3	25	12.13 (9.13–15.14)			14 (56.0)			24 (96.0)
≥3	4	5.52 (0.40–10.65)			3 (75.0)			4 (100.0)

PFS, progression-free survival; ORR, objective response rate; DCR, disease control rate; CI, confidence index; EGFR, epidermal growth factor receptor; 19del, exon 19 deletions; 21L858R, exon 21 L858R; 20ins, exon 20 insertions.

Safety

AEs of any grade occurred in 22 patients (37.9%) and mainly included diarrhea (n=12, 20.7%), rash (n=10, 17.2%), stomatitis (n=6, 10.3%), and paronychia (n=6, 10.3%). Five patients (8.6%) developed diarrhea ≥ grade 3 but all recovered after symptomatic treatment. No patients discontinued afatinib treatment due to AEs ().

Table 4

AEs in the overall population

AEs	Any, n (%)	≥ Grade 3, n (%)
Total	22 (37.9)	5 (8.6)
Diarrhea	12 (20.7)	5 (8.6)
Rash	10 (17.2)	0 (0.0)
Stomatitis	6 (10.3)	0 (0.0)
Paronychia	6 (10.3)	0 (0.0)
Fatigue	5 (8.6)	0 (0.0)
Nausea	1 (1.7)	0 (0.0)
Gingivitis	1 (1.7)	0 (0.0)

AEs, adverse events.

Discussion

The current study demonstrated that afatinib at 30 mg/d was effective for EGFR-mutated advanced LAD with a favorable PFS of 9.8 months for the overall population. For common and uncommon EGFR mutations except for 20ins, afatinib 30 mg/d as the first-line regimen was associated with a PFS of 14.5 months and comparable 10.4 months, respectively. The corresponding ORR for common and uncommon subgroups was 60% and 80%, respectively. In addition, a well-tolerated safety profile was observed for afatinib starting from 30 mg/d.

Afatinib has been routinely recommended to start from 40 mg/d based on the results of the LUX-Lung trials. However, in the LUX-Lung trials and subsequent real-world studies, more than a quarter or even up to half of patients receiving afatinib 40 mg/d initially had to reduce the dose eventually due to AEs (2,3,13-15,23). The high frequency of AEs related to afatinib at 40 mg/d not only lowers the quality of life of patients but can also cause treatment discontinuation (24). On the contrary, afatinib starting from 30 mg/d may be more appropriate, especially for Asians patients with NSCLC (15,25). Recent studies demonstrated that an initial afatinib dose <40 mg or 30 mg/d resulted in a similar response and PFS to the dose of 40 mg/d but resulted in fewer serious AEs for EGFR-mutated LAD (25,26). The incidence of diarrhea as the most commonly experienced AE was only 41% at 30 mg/d compared with 100% at 40 mg/d (25). In the current study, we showed a similar favorable safety profile. Afatinib 30 mg/d was well tolerated, with an incidence of 37.9% for any-grade AEs and only 8.6% for manageable grade 3 diarrhea. More importantly, no patient discontinued treatment due to AEs, which may contribute to prolonged survival.

Afatinib had comparable efficacy with first-generation EGFR-TKIs in the treatment of LAD with EGFR common mutations. In the LUX-Lung 3 trial, the PFS for patients with EGFR 19del and 21L858R who received first-line afatinib at 40 mg was 13.6 months (13). A similar first-line PFS for LAD patients with common EGFR mutations treated with afatinib at 30 mg was observed in the current study. A previous study from Taiwan, which intended to investigate the efficacy of afatinib starting from 30 mg, also reported a similar PFS of 469 and 443 days for 19del and 21L858R mutations, respectively (25). Furthermore, afatinib at 30 mg was not only associated with non-inferior efficacy but a better safety profile compared with the higher dose of 40 mg. The low frequency of AEs and good tolerability in the present study was consistent with findings in the abovementioned study from Taiwan, China (25).

Different EGFR-targeted drugs have shown variable efficacy for uncommon mutations within EGFR exon 18–21 (27,28). Previous studies demonstrated that patients with uncommon EGFR mutations treated with afatinib might have a better prognosis than those treated with first-generation TKIs (4,7,29,30). Afatinib at 40 mg for LAD with major uncommon EGFR mutations (G719X, L861Q, and S768I) resulted in a varied PFS of 10.7–17.1 months and an ORR around 50–74% (6,31,32). However, little evidence has been released concerning the efficacy of afatinib at 30 mg in these patients. Our findings confirmed that uncommon exon 18–21 mutations except exon 20ins were sensitive to afatinib even at 30 mg (6,26,29). It is noteworthy that real-world studies often enroll patients usually excluded from clinical trials. For example, our study included approximately one-fifth of patients with brain metastases who had a dismal PFS of merely 5 months. Furthermore, while uncommon EGFR mutations account for 10–20% of all EGFR mutations, nearly half of our population had uncommon EGFR mutations. Afatinib applied as the first-line treatment was associated with a favorable PFS, which we speculate was linked to improved tolerance from the adjusted initiation dose and promising efficacy for major uncommon mutations as well. EGFR exon 20ins are generally associated with de novo resistance to first- or second-generation EGFR-TKIs (6,33). Our result confirmed this concept, with a median PFS of only 2–3 months for this special subgroup (34).

This study has a few limitations. First, the limited sample size may affect the results. Considering the small population size and genetic diversity of patients with uncommon EGFR mutations, it is difficult to collect a large enough sample and large-scale randomized trials on the current topic also seem to be infeasible. Second, we did not include EGFR T790M and other rare mutations. Third, this was an observational study without setting a control group of afatinib 40 mg/d, although previous studies have demonstrated similar efficacy of 30 mg/d compared to 40 mg/d (25,26). As an observational retrospective study, the descriptive data of efficacy and safety profile, although partly in line with previous studies, is necessary to confirm in the future large-scale studies.

In conclusion, our findings confirmed previous findings that afatinib was not only effective for LAD with common EGFR mutations but potent for those with EGFR G719X/S768I/L861Q and classical compound mutations. More importantly, a starting dose of afatinib 30 mg instead of 40 mg was used in this study and favorable toxicity was observed. Taken together, we believe that afatinib at 30 mg/d may be an option for the treatment of EGFR-mutated LAD especially for those with major uncommon EGFR mutations.

The article’s supplementary files as