- Email: [email protected]
The EGFR mutation and its correlation with response of gefitinib in previously treated Chinese patients with advanced non-small-cell lung cancer
Original article
Annals of Oncology 16: 1334– 1342, 2005 doi:10.1093/annonc/mdi340 Published online 14 June 2005
The EGFR mutation and its correlation with response of gefitinib in previously treated Chinese patients with advanced non-small-cell lung cancer X.-T. Zhang1*, L.-Y. Li1, X.-L. Mu1, Q.-C. Cui2, X.-Y. Chang2, W. Song3, S.-L. Wang1, M.-Z. Wang1, W. Zhong1 & L. Zhang1 1
Received 9 March 2005; revised 1 May 2005; accepted 3 May 2005
Background: The aim of the study was to evaluate the efficacy of gefitinib and the epidermal growth factor receptor (EGFR) mutation to gefitinib response in a series of Chinese patients with pretreated advanced non-small-cell lung cancer (NSCLC). Methods: A total of 98 patients who had failed at least one platinum-based regimen received gefitinib 250 mg once daily. The mutation analysis of the EGFR kinase domain was performed for 30 patients using paraffin-embedded tumor tissue. Results: The response rate was 31.6% and the disease control rate was 67.3%. Objective response was correlated with adenocarcinoma, female gender and non-smokers. Median progress free survival (PFS) was 7.0 months, median overall survival (OS) was 12.0 months and 1-year survival was 53.1%. The median PFS and OS were improved among patients with adenocarcinoma, gefitinib responders and non-smokers. Active gene mutation was detected in 12 patients. Mutation rates were higher among gefitinib responders, non-smokers, patients with adenocarcinoma and female patients. OS was longer for patients with gene mutation than for patients without mutation. Conclusion: Gefitinib demonstrated significant antitumor activity with a favorable toxicity profile for pretreated Chinese patients with advanced NSCLC. The active mutation of the EGFR kinase domain was strongly associated with response to gefitinib and prolonged overall survival. Key words: EGFR mutation, gefitinib, non-small-cell lung cancer, p-EGFR, predictive marker, targeted therapy
Introduction Lung cancer is the most common cause of cancer death in both men and women worldwide. The majority of these deaths are due to non-small-cell lung cancer (NSCLC) [1]. Despite advances in treatment such as combination chemotherapy and chemoradiotherapy, the 5-year survival rate for all stages of lung cancer is 12%. The median survival time for patients with advanced disease receiving cisplatin-based chemotherapy is 8–10 months. Prognosis is particular poor for patients who have progressive disease following chemotherapy, with a median survival time of 16 weeks and a 1-year survival rate of 16% [2, 3]. Therefore effective low-toxicity treatments for patients with advanced NSCLC are needed.
*Correspondence to: Dr X.-T. Zhang, Department of Pulmonary Medicine, Peking Union Medical College Hospital and Chinese Academy of Medical Sciences, Beijing 100730, China. Tel: + 86-010-84251629; Fax: + 86-010-84251629; E-mail: [email protected] q 2005 European Society for Medical Oncology
It has been clearly demonstrated that aberrant activity of epidermal growth factor receptor (EGFR) signaling plays a key role in the development of tumor cell growth in a variety of common solid tumors, including NSCLCs [4]. Once activated, tyrosine kinase can promote autophosphorylation of protein tyrosine residue, thereby switching on intracellular signal transmission. Inhibition of tyrosine kinase phosphorylation by agents designed to target these specific molecular process appears to induce substantial therapeutic benefits [5]. Gefitinib (Iressae, ZD1839) is an orally active and highly effective EGFR tyrosine kinase selective inhibitor that blocks signal transduction pathways implicated in cancer growth [6]. Phase I studies showed that gefitinib was generally well tolerated, with evidence of tumor regression in patients with NSCLC [7, 8]. Two large phase II studies indicated clinically meaningful antitumor activity in previously treated patients with advanced NSCLC [9, 10]. Although EGFR is commonly expressed in NSCLC, there was no evidence for a consistent relationship between EGFR expression levels and response to
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Department of Pulmonary Medicine, 2Department of Pathology, 3Department of Radiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China
1335 gefitinib [11]. Recently, two studies have shown that a gainof-function somatic mutation of EGFR accounts for the response of some patients with NSCLC to gefitinib [12, 13]. The efficacy of gefitinib has been explored extensively in the West and in Japan. We aimed to assess the antitumor activity of gefitinib as well as the relationship between response and EGFR mutation in a series of Chinese patients in Peking Union Medical College Hospital.
Patients and methods Patient eligibility
Drug administration One tablet of gefitinib (250 mg) was taken once daily at about the same time. Patients continued the course uninterrupted until disease progression, intolerable toxicity or withdrawal of consent. All drugs were supplied by AstraZeneca through the Extensive Access Program and local trail. No systemic anticancer treatment was permitted during the study, except for palliative radiotherapy in patients with isolated symptomatic bone metastases provided that drug administration was not interrupted for >14 days.
Assessment of response and toxicity Baseline evaluation included medical history and physical examination, electrocardiogram, chest radiography, thorax CT scan and ultrasonography of the upper abdomen. Laboratory investigations included complete blood counts, urinalysis and renal and liver function tests. Performance status was evaluated according to the Eastern Cooperative Oncology Group (ECOG) criteria. MRI of the brain and radionuclide bone scans were performed only if metastatic disease was suspected according to the clinical manifestations of each patient. Patients were re-evaluated using the same method as at the baseline at the end of the first and third months of therapy, and then every 3 months. Objective tumor response and its duration were assessed according to the RECIST criteria [14], and all _ 28 days after the initial assessment of responses had to be confirmed > response. Progress-free survival (PFS) was calculated from the date of start of therapy to the date of disease progression or last follow-up. Overall survival (OS) was assessed from the first dose of gefitinib to the date of death for any reason. Patients alive at the data cut-off were censored at the last date that the patient was known to be alive. All patients were evaluated in an intent-to-treat analysis. All adverse events were recorded and graded according to National Cancer Institute Common Toxicity Criteria, Version 2.0 [15].
EGFR gene mutation analysis Tumor tissues obtained before any systemic treatment were collected retrospectively from patients treated with gefitinib. All tumor tissues were formalin-fixed and embedded in paraffin. DNA was extracted from
Immunoactivity analysis Phosphorylation EGFR (p-EGFR) status was evaluated by immunohistochemical staining using Moab phospho-EGFr Tyr1068 (Cell Signaling, Beverly, MA) according to the manufacturer’s instruction. The slides were evaluated by two independent pathologists. The result of membrane and cytoplasma immunoreactivity was scored based on the percentage of stained cells and staining intensity according to a three-stage grading system: 0, no tumor cells were stained; 1+ , <50% of the tumor cells were stained or weakly stained; 2+ , >50% cells were stained vividly. The negative group includes tissues with score 0, and the positive group includes tissues with score 1+ to 2+ .
Statistical analysis Logistical regression test models were used to identify baseline factors (gender, smoking history, physical status, histology, TNM stage and prior chemotherapy) that might independently predict tumor response. Median PFS and OS were calculated using the Kaplan– Meier method. The logrank test was used to detect differences in PFS and OS between strata. The gene mutation rate was compared between patients with and without objective tumor response using Pearson’s x2-test or Fisher’s exact test.
Results Patient characteristics Ninety-eight consecutive patients were enrolled in the study between October 2002 and February 2004. Patient characteristics are listed in Table 1. Seventy-six patients (77.6%) had adenocarcinoma, of whom 16 (16.3%) had bronchioloalveolar carcinoma (BAC) or adenocarcinoma with BAC features. All patients had received at least one cycle of platinum-based chemotherapy. Fifty-eight (59.2%) patients had received one prior regimen and 40 (40.8%) patients had received two and more prior regimens. Previous chemotherapy regimens were as follows: 40.8% patients had received gemcitabine, 28.9% had received novelbine, and 57.9% and 17.1% patients had received taxol- or docetaxel-based regimens, respectively. Disease progress during previous treatment was as follows: 9.2% patients had a partial response, 42.1% patients achieved stable disease and 48.7% patients experienced disease progress.
Toxicity All patients were assessed for toxicity. Treatment-related toxicities are listed in Table 2. The most frequently reported
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
_ 18 years with histologically or cytologically confirmed Patients aged > NSCLC who had failed to respond to conventional treatment, such as chemotherapy, radiotherapy or biotherapy, were enrolled in this study after signing an informed consent form. The former treatment should have _ 28 days before this study. Other eligibility criteria included adeended > _ 60 mmHg. Excluquate bone marrow, liver and renal function and PaO2 > sion criteria included a serious pre-existing medical condition such as uncontrolled infection, interstitial pneumonia or pulmonary fibrosis, severe chronic diarrhea, pregnancy or lactation.
the tumor sections with DEXPAT (Takara Biomedical, Shiga, Japan) according to the manufacturer’s instruction. For mutational analysis of the kinase domain of EGFR coding sequence, exons 18, 19, 20 and 21 were amplified with four pairs of primers. Forward and reverse sequencing reactions were performed with the same primers for PCR amplification and ABI BigDye Terminator kit v3.1 (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. Sequencing reactions were electrophoresed on an ABI3700 genetic analyzer. Sequence variations were determined using Seqscape software (Applied Biosystems) with the EGFR reference sequence (NM-005228.3, National Center for Biotechnology Information, Hyattsville, MD). All the sequence variations were confirmed by multiple independent PCR amplifications and repeated sequencing reactions.
1336 Table 1. Patient characteristics (n = 98) No. of patients Median age
Percentage
and exacerbation of dysphasia. The patient with squamous cell carcinoma had a poor ECOG performance status of 3 and mediastinal lymph node metastases before treatment.
58 (28–85) years
Gender 58
59.2
Female
40
40.8
Non-smokers
60
61.2
Former smokersa
38
38.8
0/1
68
69.4
2
22
22.4
8
8.1
Squamous
17
17.3
Adenocarcinoma
76
77.6
5
5.1
IIIA
2
2.1
IIIB
10
10.2
IV
86
87.8
1 cycle
58
59.2
2 cycles
30
30.6
> _ 3 cycles
10
10.2
Prior radiotherapy
35
34.3
Smoking history
ECOG performance status
3/4 Histology
Unspecified TNM staging
Prior chemotherapy
a _ 10 years who Patients who smoked one pack of cigarettes daily for > stopped smoking within 6 months when lung cancer was diagnosed.
adverse events during treatment were diarrhea (40.9%) and skin disorders (60.2%), including dry skin and acne-like rash. The majority of these events were mild at CTC grade 1–2. Other adverse events included nausea (6.1%), oral ulcer (3.1%) and elevation in hepatic enzymes (2.0%). Five patients reported drug-related CTC grade 3 and 4 adverse events. Two cases of diarrhea recovered completely during treatment. Two cases of elevation in hepatic enzymes recovered after temporary interruption of the drug. One patient withdrew from the trial on day 50 of the treatment because of grade 3 skin rash
Response and survival In the intention-to-treat analysis of response rate, all patients were included whether or not the response could be properly evaluated. One patient (1%) achieved complete response and 30 patients (30.6%) achieved partial response, with an overall objective response rate (RR) of 31.6% [95% confidence interval (CI) 30.7– 32.5]. Stable disease was documented in 35 patients (35.7%) and the overall disease control rate (DCR) was 67.3% (95% CI 31.0– 32.2). Two of the 98 patients enrolled in the study died of deterioration of disease before efficacy assessment. Both patients had squamous carcinoma with a poor performance status of 3. No improvement was found in either patient during gefitinib therapy. One patient received the study drug for 50 days and was evaluated as stable disease at the first follow-up, but then withdrew because of the adverse events described above. Multivariate logistic analysis showed that the RR for the 95 evaluable patients was well associated with the histology type, gender and smoking status. Factors associated with response are shown in Table 3. The odds of objective response were 13 times higher for patients with adenocarcinoma than for patients with non-adenocardinoma [odds ratio (OR) 13.386, 95% CI 1.701– 105.316, P = 0.001], four times higher for female than for male patients (OR 4.310, 95% CI 1.733–10.752, P = 0.001) and four times higher for non-smokers than for former smokers (OR 3.914, 95% CI 1.416 –10.822, P = 0.005). Other baseline factors (age, TNM stage, physical status, prior chemotherapy) were not correlated with the objective tumor response. The median PFS was 7.0 ± 0.82 months (95%CI 5.39 – 8.61). The median OS was 12.0 ± 1.87 months (95%CI 8.33– 15.67). The 1-year survival rate was 53.1% and the 18-month survival rate was 19.4%. Median PFS and OS among patients with adenocarcinoma or non-smokers were significantly longer than those among patients with non-adenocarcinoma or former smokers (log rank test P-values of 0.000 and 0.0002, and 0.002 and 0.0035, respectively). Median PFS and OS also improved among patients who had objective tumor response or patients with disease control (log rank test P-values of
Table 2. Treatment-related toxicity (n = 98) Adverse event
Grade 1/2
Grade 3/4
Total
No.
Percentage
No.
Percentage
No.
Percentage
Dry skin
19
19.4
0
0
19
19.4
Acne-like rash
39
39.8
1
1.0
40
40.8
Diarrhea
38
38.8
2
2.0
40
40.8
Nausea
6
6.1
0
0
6
6.1
Oral ulcer
3
3.1
0
0
3
3.1
Elevation of hepatic enzymes
0
0
2
2.0
2
2.0
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Male
3.0
9.5 PR
SD ADC
ADC Never F
F
11
12
Never
13.4 PR BAC F 10
Never
11.2
8.0 PR
PR BAC
ADC F 9
Never
M 8
Never
5.8
12.4 PR
SD ADC
ADC
F
Never
F
Never
9.8
9.1 PR
PR ADC
ADC Never
3.1
9.6 PR
SD SCC
ADC
3.2 SD
Response Pathol.
ADC
7
0.0002
6
13.87
Never
8.00 ± 1.31 (5.43–10.57)
F
Non-responders
F
16.00 ± 1.59 (12.89– 19.11)
5
Responders
0.0002
4
6.00 ± 1.11 (3.82–8.18)
Never
16.00 ± 1.70 (12.67– 19.33) 14.06
Nonadenocarcinoma
Never
Adenocarcinoma
M
0.0035
8.00 ± 1.00 (6.03–9.97)
F
16.00 ± 1.76 (12.54– 19.46) 8.51
Former smokers
3
Non-smokers
2
0.2290
Former
1.45
M
16.00 ± 1.71 (12.65– 19.35)
1
9.00 ± 2.28 (4.52–13.48)
Female
P-value
Smoking status
Male
Log-rank
Gender
M ± SE (95% CI)
No.
Table 5. Factors associated with overall survival
Table 6. Clinical features and mutation, p-EGFR status (n = 12)
Only half the patients consented to providing tissue for DNA analysis, and some of the tissue used was obtained from biopsy and was insufficient for DNA extraction and analysis. Therefore EGFR gene sequencing was performed in 30 patients of whom 16 (53.3%) were male and 14 (46.7%) were female. Twenty-one (70.0%) of these patients had adenocarcinoma and nine (30.0%) had non-adenocarcinoma, 18 (60.0%) were non-smokers and 12 (40.0%) were former smokers. Nine patients showed response to gefitinib, 12 patients had stable disease and nine showed progressive disease.
PFS (months)
EGFR mutation analysis
F, female; M, male; Pathol., pathological diagnosis; ADC, adenocarcinoma; BAC, brochioloalveolar carcinoma; SCC, squamous cell carcinoma; SD, stable disease; PR, partial response; PFS, progress-free survival; OS, overall survival; D, dead; A, alive; p-EGFR, phosphorylation epidermal growth factor receptor.
–
–
L858R
L858R
2819T!G
2819T ! G 21 D
A 9.5
9.8
21
+
+ L858R 2819T!G 21 A 18.5
2819T ! G
– L858R 2819T ! G
A
21
L858R
L858R
21
+
–
L858R 2819T!G
+ L858R
+ L747–P753, insS
2819T ! G
+
–
2240–2257 del
OS (months)
0.0006 and 0.0002). The factors associated with PFS and OS are shown in Tables 4 and 5.
A
0.0006
13.7
11.63
25.3
3.00 ± 0.41 (2.19–3.81)
21
Non-responders
21
10.00 ± 1.5 (7.09–12.91)
0.0000
D
Responders
27.41
A
3.00 ± 0.32 (2.37–3.63)
16.5
8.00 ± 1.13 (5.78–10.22)
Nonadenocarcinoma
0.0002
12.4
Adenocarcinoma
14.28
21
3.00 ± 0.61 (1.81–4.19)
19
9.00 ± 1.63 (5.81–12.19)
Former smokers
0.3653
A
Non-smokers
0.82
D
8.00 ± 0.88 (6.28–9.72)
17.9
6.00 ± 1.34 (3.37–8.63)
Female
P-value
Site (exon)
Male
Log-rank
Survival status
M ± SE (95% CI)
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Table 4. Factors associated with progress-free survival
18.7
31/37
E746–A750
33/58
6/37
E746–A750
25/58
Former smokers
2236–2250 del
Non-smokers
2236–2250 del
20/20
19
0/20
19
Nonadenocarcinoma
D
0.005
44/75
A
7.429
19/39
31/75
4.6
0.000
20/39
Adenocarcinoma
14.0
12.271
Female
–
0.001
E746–A750
10.469
2235–2249 del
45/56
19
11/56
D
P-value
24.8
T
Amino acid sequence
Nonresponders
Nucleotide sequence
Male
Responders
p-EGFR status
Table 3. Factors associated with response
2819T!G
1337
1338
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Figure 1. EGFR kinase domain mutation. (A) Four patterns of nucleotide sequence change. (B, C) Nucleotide sequence with heterozygous in-frame deletion (two peaks). Tracking in both sense and antisense directions is shown to demonstrate the two breakpoints of the deletion. (D) Heterozygous missense mutations (two peaks) at nucleotide 2189 (T–G) resulting in amino acid substitutions. The corresponding wild-type sequence is shown for comparison.
1339
EGFR phosphorylation analysis Immunohistochemical analyses were performed on 45 tumors including 30 tumors with EGFR mutation analysis. The results of a representative case are shown in Figure 3. The activation EGFR is reflected by its tyrosine-phosphorylating status. The p-EGFR status was correlated with the response to gefitinib. The odds of the respone rate were higher for patients with positive p-EGFR than for those with negative p-EGFR (OR Table 7. Factors associated with gene mutations (n = 30) Mutation
Non-mutation
Male
3/16
13/16
Female
9/14
5/14
Adenocarcinoma
11/21
10/21
Nonadenocarcinoma
1/9
8/9
Non-smokers
11/18
7/18
Former smokers
1/12
11/12
Responders
8/9
1/9
Non-responders
4/21
17/21
Survival functions
1.2
0 = non, 1=mutation
1.0
1.00 1.00-censored
Cum survival
0.8
0.00
0.6
0.00-censored
0.4 0.2 0.0 0.2 0
2
4
6
8
10
12
14
PFS Survival functions 1.2
0 = non, 1=mutation 1.00
Cum survival
1.0
1.00-censored 0.00
0.8
0.00-censored
0.6 0.4 0.2 0
0
10
20
30
OS Figure 2. Survival curve for patients with or without mutation. (A) Kaplan–Meier plots showing progress-free survival (PFS) according to EGFR mutation (n = 30). (B) Kaplan–Meier plots showing overall survival (OS) according to EGFR mutation (n = 30).
2.071; 95% CI 0.955–4.491; P = 0.013). There was a tendency for prolonged PFS and OS, but this was not statistically significant. The RR, PFS and OS for patients with positive and negative p-EGFR are shown in Table 8. Six of 12 tumors with the EGFR mutation, but only four of the 18 tumors without this mutation, were p-EGFR positive.
Discussion
x
P-value
6.451
0.014
4.471
0.040
8.356
0.005
12.804
0.001
The outcome for patients with NSCLC who fail to respond or have progressive disease after first-line chemotherapy is poor. Docetaxel is considered the standard second-line chemotherapy for NSCLC. Alimtaw has recently been approved by the US Food and Drug Administration (FDA) for patients with NSCLC in whom platinum-based chemotherapy has failed. However, the therapeutic efficacy of these drugs is limited [16, 17]. Erlotinib has shown survival benefit for patients who previously failed at least one prior chemotherapy [18]. In two large phase II trials (IDEAL1 and IDEAL 2), following platinum-containing therapies, gefitinib still has
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Mutations of EGFR were detected in 12 tumors. The clinical features and the results of mutational study are listed in Table 6. Four were in-frame deletions in exon 19 (delE746A750, delL747-P753 insS). Eight mutations were in exon 21. All were the 2819T ! G, which resulted in an L858 change. Representative cases are shown in Figure 1. Eight of the nine patients who responded to gefitinib had the EGFR gene mutation in contrast with none of the nine patients with disease progression. Four of the 12 patients with disease control had the mutation. All except one of these patients had adenocarcinoma, and only one was a former smoker. The difference between the mutation rates in responders and non-responders is statistically significant (P = 0.001). Mutation rates were higher among non-smokers (11/18) than among former smokers (1/12) (P = 0.005). Eleven of 21 patients with adenocarcinoma, but only one of nine patients with non-adenocarcinoma, had a gene mutation (P = 0.040). The mutation rate was also higher among female patients (9/14) than among male patients (3/16) (P = 0.014). The factors associated with the mutation are shown in Table 7. The median PFS for patients with the mutation was 10.0 ± 1.2 months (95% CI 7.6–12.4), whereas it was 3.0 ± 1.1 months (95% CI 0.9 –5.2) for patients without the mutation. The median OS for patients without the mutation was 7.0 ± 2.0 months (95% CI 3.1 –10.9), but has not yet been reached for patients with the mutation. Nine of 12 patients are still alive with a median follow-up of 17.5 (4.6 – 25.3) months. The PFS and OS for patients with the mutation were significantly longer than for patients without the mutation (P-values of 0.0454 and 0.0022, respectively). The PFS and OS for patients with and without the mutation are shown in Figure 2.
1340
an objective remission rate of 12 –18% with a disease control rate of about 50%. On the basis of these two studies, gefitinib has been approved for the treatment of patients with NSCLC who have failed to respond to other chemotherapies [19]. In the present study, the response rate was 31.6% and the disease control rate was 67.3%, both of which are higher than in the phase II trials described above but similar to results obtained in Japan. A total of 102 Japanese patients with advanced lung cancer were evaluated in IDEAL1. Similar to our result, the RR for Japanese patients was higher than that of non-Japanese patients (27.5% compared with 10.45%; OR 3.72; P = 0023). However, the difference in RR was imputed
Table 8. Response rate, progress-free survival and overall survival among patients with positive p-EGFR and negative p-EGFR Positive p-EGFR
Negative p-EGFR
x/log rank
P-value
Responders
6/10
4/10
7.305
0.013
Non-responders
6/35
29/35
PFS [M ± SE (95% CI)]
9.80 ± 2.07 (5.75 –13.85)
5.70 ± 1.87 (2.04–9.36)
3.06
0.0805
OS [M ± SE (95% CI)]
18.70 ± 2.32 (14.16–23.24)
9.80 ± 2.17 (5.54–14.06)
3.76
0.0526
RR, response rate; PFS, progress free survival; OS, overall survival; p-EGFR, phosphorylation epidermal growth factor receptor.
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Figure 3. Immunohistochemical staining with p-EGFR antibody (magnification 200).
to bias of the baseline predictive factors of the patients (i.e. gender, PS and pathology) rather than ethnicity [9, 20]. In our study, logistic regression analysis showed that histology, gender and smoking history were prognostic factors associated with an objective tumor response. Physical status, prechemotherapy and TNM stage were not predictive factors for tumor response. The data were similar to the two phase II trials. The present study demonstrated that a daily oral dose of 250 mg of gefitinib was well tolerated in Chinese patients. Similar to prior clinical trials evaluating the toxicity of gefitinib, the most frequently reported adverse events were acne-like rash and grade 1 and 2 diarrhea. Severe adverse events were rare. It has been reported that 291 (1.7%) of 17 500 patients treated with ZD1839 in Japan had developed suspected interstitial pneumonia or acute lung injury in contrast with a lower incidence worldwide [21]. We paid more attention to this severe adverse event in our study because of the similarity in ethnicity between Chinese and Japanese, but we did not observe any evidence of interstitial pneumonia. The most recent study showed that EGFR gene mutations were correlated with response to gefitinib in NSCLC. All these mutations were within exons 18 –21 of the kinase domain of EGFR. The mutations were most frequently found in patients who responded to gefitinib, patients with adenocarcinoma, patients who were non-smokers and female patients [12, 13]. In the present study, a total of 12 mutations were detected, with four in-frame deletion mutations in exon 19 and eight missense mutations in exon 21. The mutation rate was significantly higher for gefitinib responders, non-smokers, female patients and patients with adenocarcinoma than for gefitinib non-responders, male patients, former smokers and patients with non-adenocarcinoma. These data agreed with the above studies and clinical data. The previous study showed that the gene mutation rate of Japanese (26%) was much higher than that of non-Japanese. The mutation rate of Japanese patients with adenocarcinoma was 46% [12]. That may partially explain the different response rates of Japanese and Caucasians. The therapeutic efficacy for Chinese patients in the present study was close to that for Japanese which suggests that the EGFR gene mutation rate would be higher in the Chinese population because of the ethnic similarity. EGFR is frequently expressed in NSCLC, and EGFR activation after ligand binding is believed to be its most important
1341
Conclusion In conclusion, a daily oral dose of gefitinib 250 mg is well tolerated in Chinese patients with advanced NSCLC and provides significant antitumor activity in patients who have failed to respond to previous platinum-based chemotherapy. The gene mutation rate of EGFR kinase domain was much higher among gefitinib responders than among non-responders.
Patients with mutation have a longer OS than patients without mutation. Therefore gefitinib can be used as second- or thirdline therapy for treatment of Chinese patients with advanced NSCLC, and gene mutation of the EGFR kinase domain can be used as a predictive marker for patient selection.
References 1. Lung Cancer Collaborative Group. Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised clinical trails. BMJ 1995; 311: 899–909. 2. Schill J. Current standards of care in small-cell and non-small-cell lung cancer. Oncology 2001; 61: 3– 13. 3. Haura EB. Treatment of advanced non-small-cell lung cancer: a review of current randomized clinical trials and an examination of emerging therapies. Cancer Control 2001; 8: 326–336. 4. Moscatello DK, Holgado-Madruga M, Godwin AK et al. Frequent expression of a mutant epidermal growth factor receptor in multiple human tumors. Cancer Res 1995; 55: 5536–5539. 5. Woodburn JR. The epidermal growth factor receptor and its inhibition in cancer therapy. Pharmacol Ther 1999; 82: 241 –250. 6. Baselga J. Why the epidermal growth factor receptor? The rationale for cancer therapy. Oncologist 2002; 7 (Suppl 4): 2–8. 7. Ranson M, Hammond LA, Ferry D et al. ZD1839, a selective oral epidermal growth factor receptor-tyrosine kinase inhibitor, is well tolerated and active in patients with solid, malignant tumors: results of a phase I trial. J. Clin Oncol 2002; 20: 2240–2250. 8. Baselga J, Rischin D, Ranson M et al. Phase I safety, pharmacokinetic, and pharmacodynamic trial of ZD1839, a selective oral epidermal growth factor receptor tyrosine kinase inhibitor, in patients with five selected solid tumor types. J Clin Oncol 2002; 20: 4292–4302. 9. Fukuoka M, Yano S, Giaccone G. Multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small-cell lung cancer. J Clin Oncol 2003; 21: 2237–2246. 10. Kris MG, Natale RB, Herst RS et al. Efficacy of gefitinib, an inhibitor of the epidermal growth factor tyrosine kinase, in symptomatic patients with platinum-based chemotherapy. J Clin Oncol 2000; 18: 2095–2103. 11. Bailey R, Kris M, Wolf M et al. Gefitinib (‘Iressa’, ZD 1839) monotherapy for pretreated advanced non-small cell lung cancer in IDEAL 1 and 2: tumor response is not clinically relevantly predictable from tumor EGFR membrane staining alone. Lung Cancer 2003; 41 (Suppl 571): (Abstr). 12. Paez JG, Janne PA, Lee JC et al. EGFR mutation in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004; 304: 1497–1500. 13. Lynch TJ, Bell DW, Sordella R et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of nonsmall-cell lung cancer to gefitinib. N Engl J Med 2004; 350: 2129–2139. 14. Response evaluation criteria in solid tumors (RECIST) quick reference. Available online at: http://www.nci.nih.gov/bip/RECIST.htm. 15. Cancer Therapy Evaluation Program, Common Toxicity Criteria, Version 2.0, DCTD, NCI, NIH, DHHS. Available online at: http://ctep. cancer.gov. 16. Shepheral FA, Dancey J, Ramlan R et al. Prospective randomized trial of docetaxel versus best supportive care in patients with non-small-cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol 2000; 18: 2095–2103.
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
role in clinical settings. The relationships of EGFR expression and EGFR activation status to the sensitivity of EGFR inhibitors have been studied in a few retrospective series. The predictive association of EGFR and p-EGFR with tumor sensitivity has not been verified, but the activation of the downstream signaling molecule (p-Akt) seems to be highly sensitive to gefitinib in NSCLC [22–25]. In the present study, patients with positive p-EGFR expression had a higher response rate (P = 0.013) and a tendency to have longer median PFS and OS (P-values of 0.0805 and 0.0526, respectively). The relationships between p-EGFR activation and activation of the downstream signaling pathways, as well as the relationship of activation status about EGFR and downstream signaling pathways to the response of gefitinib need further investigation. In the present study, EGFR phosphorylation was detected by immunohistochemical analysis in six out of 12 tumors with EGFR mutation in contrast with four out of 19 tumors without EGFR mutation. The data agreed with the results obtained by Huang et al. [26]. In their study, increased EGFR phosphorylation was detected by immunoblot in 26 out of 39 tumors with EGFR mutation compared with 14 out of 38 tumors without mutation. In general, the EGFR kinase domain mutations seem to have a different effect on kinase activation. The functional significance of mutation on phosphorylation requires additional investigation. Although previous studies have shown that the EGFR kinase mutations are correlated with clinical response, it is possible that not all patients with EGFR mutation respond to gefitinib. In the present study, four patients with disease control also had EGFR gene mutation, which has not been documented in other studies. All except one of these four patients had histological evidence of adenocarcinoma and OS was >10 months (follow-up 10– 25 months). Our data suggest that even patients with gene mutation who failed to show objective tumor response could probably also benefit from gefitinib treatment. In general, OS for patients with mutation was significant longer than that for patients without mutation. Further study of the antitumor mechanism of gefitinib may provide better selection criteria for patients. We should note that all mutations were performed with tissues at diagnosis. Could this gene status represent that after exposure to gefitinib? Does cytotoxic therapy affect the mutation status? At present, we lack information on the incidence of mutation after cytotoxic therapy. Further investigations are needed to determine the true incidence in untreated and treated tumors.
1342
23.
24.
25.
26.
the 94th Annual Meeting of the American Association for Cancer Research. Washington, DC; July 2003. Philadelphia, PA: AACR, 1362. Cappuzzo F, Gregorc V, Rossi E et al. Gefitinib in pretreated nonsmall cell lung cancer (NSCLC): Analysis of efficacy and correlation with HER2 and epidermal growth factor receptor expression in locally advanced or metastatic NSCLC. J Clin Oncol 2003; 21: 2658–2663. Han S-W, Hwang PG, Chung DH et al. Epidermal growth factor receptor (EGFR) downstream molecules as response predictive markers for gefitinib (Iressa, ZD 1839) in chemotherapy-resistant non small cell lung cancer. Int J Cancer 2005; 113: 109– 115. Cappuzzo F, Magrini E, Cereoli GL et al. Akt phosphorylation and gefitinib effficacy in patients with advanced non-small cell lung cancer. Int J Cancer 2004; 96: 1133–1141. Huang SF, Liu HP, Li LH et al. High frequency of epidermal growth factor receptor mutations with complex patterns in non-small cell lung cancer related to gefitinib responsiveness in Taiwan. Clin Cancer Research 2004; 10: 8195–8233.
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
17. De Marinis F, De Petris L. Pemetrexed in second-line treatment of non-small-cell lung cancer. Oncology (Huntingt) 2004; 18 (Suppl 8): 38 –42. 18. New drug to treat most common lung cancer. FDA Consum 2005; 39: 3. 19. Cohen MH, Williams GA, Sridhara R et al. FDA drug approval summary: gefitinib (ZD1839) (Iressa) tablets. Oncologist 2003; 8: 303 –306. 20. Nakagawa K, Tamura T, Negoro S et al. Phase I pharmacokinetic trial of the selective oral epidermal growth factor receptor tyrosine kinase inhibitor gefitinib (‘Iressa’, ZD1839) in Japanese patients with solid malignant tumors. Ann Oncol 2003; 14: 922–930. 21. Inoue A, Saijo Y, Maemondo M, Gomi K et al. Severe acute interstitial pneumonia and gefitinib. Lancet 2003; 361: 137–139. 22. Bailey LR, Kris M, Wolf M et al. Tumor EGFR membrane staining is not clinically relevant for predicting response in patients receiving gifitinib (Iressa, ZD 1839) monotherapy for pretreated advanced non-smallcell lung cancer: IDEAL 1 and 2 (abstract no. LB-170). Proceedings of
Annals of Oncology 16: 1334– 1342, 2005 doi:10.1093/annonc/mdi340 Published online 14 June 2005
The EGFR mutation and its correlation with response of gefitinib in previously treated Chinese patients with advanced non-small-cell lung cancer X.-T. Zhang1*, L.-Y. Li1, X.-L. Mu1, Q.-C. Cui2, X.-Y. Chang2, W. Song3, S.-L. Wang1, M.-Z. Wang1, W. Zhong1 & L. Zhang1 1
Received 9 March 2005; revised 1 May 2005; accepted 3 May 2005
Background: The aim of the study was to evaluate the efficacy of gefitinib and the epidermal growth factor receptor (EGFR) mutation to gefitinib response in a series of Chinese patients with pretreated advanced non-small-cell lung cancer (NSCLC). Methods: A total of 98 patients who had failed at least one platinum-based regimen received gefitinib 250 mg once daily. The mutation analysis of the EGFR kinase domain was performed for 30 patients using paraffin-embedded tumor tissue. Results: The response rate was 31.6% and the disease control rate was 67.3%. Objective response was correlated with adenocarcinoma, female gender and non-smokers. Median progress free survival (PFS) was 7.0 months, median overall survival (OS) was 12.0 months and 1-year survival was 53.1%. The median PFS and OS were improved among patients with adenocarcinoma, gefitinib responders and non-smokers. Active gene mutation was detected in 12 patients. Mutation rates were higher among gefitinib responders, non-smokers, patients with adenocarcinoma and female patients. OS was longer for patients with gene mutation than for patients without mutation. Conclusion: Gefitinib demonstrated significant antitumor activity with a favorable toxicity profile for pretreated Chinese patients with advanced NSCLC. The active mutation of the EGFR kinase domain was strongly associated with response to gefitinib and prolonged overall survival. Key words: EGFR mutation, gefitinib, non-small-cell lung cancer, p-EGFR, predictive marker, targeted therapy
Introduction Lung cancer is the most common cause of cancer death in both men and women worldwide. The majority of these deaths are due to non-small-cell lung cancer (NSCLC) [1]. Despite advances in treatment such as combination chemotherapy and chemoradiotherapy, the 5-year survival rate for all stages of lung cancer is 12%. The median survival time for patients with advanced disease receiving cisplatin-based chemotherapy is 8–10 months. Prognosis is particular poor for patients who have progressive disease following chemotherapy, with a median survival time of 16 weeks and a 1-year survival rate of 16% [2, 3]. Therefore effective low-toxicity treatments for patients with advanced NSCLC are needed.
*Correspondence to: Dr X.-T. Zhang, Department of Pulmonary Medicine, Peking Union Medical College Hospital and Chinese Academy of Medical Sciences, Beijing 100730, China. Tel: + 86-010-84251629; Fax: + 86-010-84251629; E-mail: [email protected] q 2005 European Society for Medical Oncology
It has been clearly demonstrated that aberrant activity of epidermal growth factor receptor (EGFR) signaling plays a key role in the development of tumor cell growth in a variety of common solid tumors, including NSCLCs [4]. Once activated, tyrosine kinase can promote autophosphorylation of protein tyrosine residue, thereby switching on intracellular signal transmission. Inhibition of tyrosine kinase phosphorylation by agents designed to target these specific molecular process appears to induce substantial therapeutic benefits [5]. Gefitinib (Iressae, ZD1839) is an orally active and highly effective EGFR tyrosine kinase selective inhibitor that blocks signal transduction pathways implicated in cancer growth [6]. Phase I studies showed that gefitinib was generally well tolerated, with evidence of tumor regression in patients with NSCLC [7, 8]. Two large phase II studies indicated clinically meaningful antitumor activity in previously treated patients with advanced NSCLC [9, 10]. Although EGFR is commonly expressed in NSCLC, there was no evidence for a consistent relationship between EGFR expression levels and response to
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Department of Pulmonary Medicine, 2Department of Pathology, 3Department of Radiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China
1335 gefitinib [11]. Recently, two studies have shown that a gainof-function somatic mutation of EGFR accounts for the response of some patients with NSCLC to gefitinib [12, 13]. The efficacy of gefitinib has been explored extensively in the West and in Japan. We aimed to assess the antitumor activity of gefitinib as well as the relationship between response and EGFR mutation in a series of Chinese patients in Peking Union Medical College Hospital.
Patients and methods Patient eligibility
Drug administration One tablet of gefitinib (250 mg) was taken once daily at about the same time. Patients continued the course uninterrupted until disease progression, intolerable toxicity or withdrawal of consent. All drugs were supplied by AstraZeneca through the Extensive Access Program and local trail. No systemic anticancer treatment was permitted during the study, except for palliative radiotherapy in patients with isolated symptomatic bone metastases provided that drug administration was not interrupted for >14 days.
Assessment of response and toxicity Baseline evaluation included medical history and physical examination, electrocardiogram, chest radiography, thorax CT scan and ultrasonography of the upper abdomen. Laboratory investigations included complete blood counts, urinalysis and renal and liver function tests. Performance status was evaluated according to the Eastern Cooperative Oncology Group (ECOG) criteria. MRI of the brain and radionuclide bone scans were performed only if metastatic disease was suspected according to the clinical manifestations of each patient. Patients were re-evaluated using the same method as at the baseline at the end of the first and third months of therapy, and then every 3 months. Objective tumor response and its duration were assessed according to the RECIST criteria [14], and all _ 28 days after the initial assessment of responses had to be confirmed > response. Progress-free survival (PFS) was calculated from the date of start of therapy to the date of disease progression or last follow-up. Overall survival (OS) was assessed from the first dose of gefitinib to the date of death for any reason. Patients alive at the data cut-off were censored at the last date that the patient was known to be alive. All patients were evaluated in an intent-to-treat analysis. All adverse events were recorded and graded according to National Cancer Institute Common Toxicity Criteria, Version 2.0 [15].
EGFR gene mutation analysis Tumor tissues obtained before any systemic treatment were collected retrospectively from patients treated with gefitinib. All tumor tissues were formalin-fixed and embedded in paraffin. DNA was extracted from
Immunoactivity analysis Phosphorylation EGFR (p-EGFR) status was evaluated by immunohistochemical staining using Moab phospho-EGFr Tyr1068 (Cell Signaling, Beverly, MA) according to the manufacturer’s instruction. The slides were evaluated by two independent pathologists. The result of membrane and cytoplasma immunoreactivity was scored based on the percentage of stained cells and staining intensity according to a three-stage grading system: 0, no tumor cells were stained; 1+ , <50% of the tumor cells were stained or weakly stained; 2+ , >50% cells were stained vividly. The negative group includes tissues with score 0, and the positive group includes tissues with score 1+ to 2+ .
Statistical analysis Logistical regression test models were used to identify baseline factors (gender, smoking history, physical status, histology, TNM stage and prior chemotherapy) that might independently predict tumor response. Median PFS and OS were calculated using the Kaplan– Meier method. The logrank test was used to detect differences in PFS and OS between strata. The gene mutation rate was compared between patients with and without objective tumor response using Pearson’s x2-test or Fisher’s exact test.
Results Patient characteristics Ninety-eight consecutive patients were enrolled in the study between October 2002 and February 2004. Patient characteristics are listed in Table 1. Seventy-six patients (77.6%) had adenocarcinoma, of whom 16 (16.3%) had bronchioloalveolar carcinoma (BAC) or adenocarcinoma with BAC features. All patients had received at least one cycle of platinum-based chemotherapy. Fifty-eight (59.2%) patients had received one prior regimen and 40 (40.8%) patients had received two and more prior regimens. Previous chemotherapy regimens were as follows: 40.8% patients had received gemcitabine, 28.9% had received novelbine, and 57.9% and 17.1% patients had received taxol- or docetaxel-based regimens, respectively. Disease progress during previous treatment was as follows: 9.2% patients had a partial response, 42.1% patients achieved stable disease and 48.7% patients experienced disease progress.
Toxicity All patients were assessed for toxicity. Treatment-related toxicities are listed in Table 2. The most frequently reported
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
_ 18 years with histologically or cytologically confirmed Patients aged > NSCLC who had failed to respond to conventional treatment, such as chemotherapy, radiotherapy or biotherapy, were enrolled in this study after signing an informed consent form. The former treatment should have _ 28 days before this study. Other eligibility criteria included adeended > _ 60 mmHg. Excluquate bone marrow, liver and renal function and PaO2 > sion criteria included a serious pre-existing medical condition such as uncontrolled infection, interstitial pneumonia or pulmonary fibrosis, severe chronic diarrhea, pregnancy or lactation.
the tumor sections with DEXPAT (Takara Biomedical, Shiga, Japan) according to the manufacturer’s instruction. For mutational analysis of the kinase domain of EGFR coding sequence, exons 18, 19, 20 and 21 were amplified with four pairs of primers. Forward and reverse sequencing reactions were performed with the same primers for PCR amplification and ABI BigDye Terminator kit v3.1 (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. Sequencing reactions were electrophoresed on an ABI3700 genetic analyzer. Sequence variations were determined using Seqscape software (Applied Biosystems) with the EGFR reference sequence (NM-005228.3, National Center for Biotechnology Information, Hyattsville, MD). All the sequence variations were confirmed by multiple independent PCR amplifications and repeated sequencing reactions.
1336 Table 1. Patient characteristics (n = 98) No. of patients Median age
Percentage
and exacerbation of dysphasia. The patient with squamous cell carcinoma had a poor ECOG performance status of 3 and mediastinal lymph node metastases before treatment.
58 (28–85) years
Gender 58
59.2
Female
40
40.8
Non-smokers
60
61.2
Former smokersa
38
38.8
0/1
68
69.4
2
22
22.4
8
8.1
Squamous
17
17.3
Adenocarcinoma
76
77.6
5
5.1
IIIA
2
2.1
IIIB
10
10.2
IV
86
87.8
1 cycle
58
59.2
2 cycles
30
30.6
> _ 3 cycles
10
10.2
Prior radiotherapy
35
34.3
Smoking history
ECOG performance status
3/4 Histology
Unspecified TNM staging
Prior chemotherapy
a _ 10 years who Patients who smoked one pack of cigarettes daily for > stopped smoking within 6 months when lung cancer was diagnosed.
adverse events during treatment were diarrhea (40.9%) and skin disorders (60.2%), including dry skin and acne-like rash. The majority of these events were mild at CTC grade 1–2. Other adverse events included nausea (6.1%), oral ulcer (3.1%) and elevation in hepatic enzymes (2.0%). Five patients reported drug-related CTC grade 3 and 4 adverse events. Two cases of diarrhea recovered completely during treatment. Two cases of elevation in hepatic enzymes recovered after temporary interruption of the drug. One patient withdrew from the trial on day 50 of the treatment because of grade 3 skin rash
Response and survival In the intention-to-treat analysis of response rate, all patients were included whether or not the response could be properly evaluated. One patient (1%) achieved complete response and 30 patients (30.6%) achieved partial response, with an overall objective response rate (RR) of 31.6% [95% confidence interval (CI) 30.7– 32.5]. Stable disease was documented in 35 patients (35.7%) and the overall disease control rate (DCR) was 67.3% (95% CI 31.0– 32.2). Two of the 98 patients enrolled in the study died of deterioration of disease before efficacy assessment. Both patients had squamous carcinoma with a poor performance status of 3. No improvement was found in either patient during gefitinib therapy. One patient received the study drug for 50 days and was evaluated as stable disease at the first follow-up, but then withdrew because of the adverse events described above. Multivariate logistic analysis showed that the RR for the 95 evaluable patients was well associated with the histology type, gender and smoking status. Factors associated with response are shown in Table 3. The odds of objective response were 13 times higher for patients with adenocarcinoma than for patients with non-adenocardinoma [odds ratio (OR) 13.386, 95% CI 1.701– 105.316, P = 0.001], four times higher for female than for male patients (OR 4.310, 95% CI 1.733–10.752, P = 0.001) and four times higher for non-smokers than for former smokers (OR 3.914, 95% CI 1.416 –10.822, P = 0.005). Other baseline factors (age, TNM stage, physical status, prior chemotherapy) were not correlated with the objective tumor response. The median PFS was 7.0 ± 0.82 months (95%CI 5.39 – 8.61). The median OS was 12.0 ± 1.87 months (95%CI 8.33– 15.67). The 1-year survival rate was 53.1% and the 18-month survival rate was 19.4%. Median PFS and OS among patients with adenocarcinoma or non-smokers were significantly longer than those among patients with non-adenocarcinoma or former smokers (log rank test P-values of 0.000 and 0.0002, and 0.002 and 0.0035, respectively). Median PFS and OS also improved among patients who had objective tumor response or patients with disease control (log rank test P-values of
Table 2. Treatment-related toxicity (n = 98) Adverse event
Grade 1/2
Grade 3/4
Total
No.
Percentage
No.
Percentage
No.
Percentage
Dry skin
19
19.4
0
0
19
19.4
Acne-like rash
39
39.8
1
1.0
40
40.8
Diarrhea
38
38.8
2
2.0
40
40.8
Nausea
6
6.1
0
0
6
6.1
Oral ulcer
3
3.1
0
0
3
3.1
Elevation of hepatic enzymes
0
0
2
2.0
2
2.0
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Male
3.0
9.5 PR
SD ADC
ADC Never F
F
11
12
Never
13.4 PR BAC F 10
Never
11.2
8.0 PR
PR BAC
ADC F 9
Never
M 8
Never
5.8
12.4 PR
SD ADC
ADC
F
Never
F
Never
9.8
9.1 PR
PR ADC
ADC Never
3.1
9.6 PR
SD SCC
ADC
3.2 SD
Response Pathol.
ADC
7
0.0002
6
13.87
Never
8.00 ± 1.31 (5.43–10.57)
F
Non-responders
F
16.00 ± 1.59 (12.89– 19.11)
5
Responders
0.0002
4
6.00 ± 1.11 (3.82–8.18)
Never
16.00 ± 1.70 (12.67– 19.33) 14.06
Nonadenocarcinoma
Never
Adenocarcinoma
M
0.0035
8.00 ± 1.00 (6.03–9.97)
F
16.00 ± 1.76 (12.54– 19.46) 8.51
Former smokers
3
Non-smokers
2
0.2290
Former
1.45
M
16.00 ± 1.71 (12.65– 19.35)
1
9.00 ± 2.28 (4.52–13.48)
Female
P-value
Smoking status
Male
Log-rank
Gender
M ± SE (95% CI)
No.
Table 5. Factors associated with overall survival
Table 6. Clinical features and mutation, p-EGFR status (n = 12)
Only half the patients consented to providing tissue for DNA analysis, and some of the tissue used was obtained from biopsy and was insufficient for DNA extraction and analysis. Therefore EGFR gene sequencing was performed in 30 patients of whom 16 (53.3%) were male and 14 (46.7%) were female. Twenty-one (70.0%) of these patients had adenocarcinoma and nine (30.0%) had non-adenocarcinoma, 18 (60.0%) were non-smokers and 12 (40.0%) were former smokers. Nine patients showed response to gefitinib, 12 patients had stable disease and nine showed progressive disease.
PFS (months)
EGFR mutation analysis
F, female; M, male; Pathol., pathological diagnosis; ADC, adenocarcinoma; BAC, brochioloalveolar carcinoma; SCC, squamous cell carcinoma; SD, stable disease; PR, partial response; PFS, progress-free survival; OS, overall survival; D, dead; A, alive; p-EGFR, phosphorylation epidermal growth factor receptor.
–
–
L858R
L858R
2819T!G
2819T ! G 21 D
A 9.5
9.8
21
+
+ L858R 2819T!G 21 A 18.5
2819T ! G
– L858R 2819T ! G
A
21
L858R
L858R
21
+
–
L858R 2819T!G
+ L858R
+ L747–P753, insS
2819T ! G
+
–
2240–2257 del
OS (months)
0.0006 and 0.0002). The factors associated with PFS and OS are shown in Tables 4 and 5.
A
0.0006
13.7
11.63
25.3
3.00 ± 0.41 (2.19–3.81)
21
Non-responders
21
10.00 ± 1.5 (7.09–12.91)
0.0000
D
Responders
27.41
A
3.00 ± 0.32 (2.37–3.63)
16.5
8.00 ± 1.13 (5.78–10.22)
Nonadenocarcinoma
0.0002
12.4
Adenocarcinoma
14.28
21
3.00 ± 0.61 (1.81–4.19)
19
9.00 ± 1.63 (5.81–12.19)
Former smokers
0.3653
A
Non-smokers
0.82
D
8.00 ± 0.88 (6.28–9.72)
17.9
6.00 ± 1.34 (3.37–8.63)
Female
P-value
Site (exon)
Male
Log-rank
Survival status
M ± SE (95% CI)
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Table 4. Factors associated with progress-free survival
18.7
31/37
E746–A750
33/58
6/37
E746–A750
25/58
Former smokers
2236–2250 del
Non-smokers
2236–2250 del
20/20
19
0/20
19
Nonadenocarcinoma
D
0.005
44/75
A
7.429
19/39
31/75
4.6
0.000
20/39
Adenocarcinoma
14.0
12.271
Female
–
0.001
E746–A750
10.469
2235–2249 del
45/56
19
11/56
D
P-value
24.8
T
Amino acid sequence
Nonresponders
Nucleotide sequence
Male
Responders
p-EGFR status
Table 3. Factors associated with response
2819T!G
1337
1338
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Figure 1. EGFR kinase domain mutation. (A) Four patterns of nucleotide sequence change. (B, C) Nucleotide sequence with heterozygous in-frame deletion (two peaks). Tracking in both sense and antisense directions is shown to demonstrate the two breakpoints of the deletion. (D) Heterozygous missense mutations (two peaks) at nucleotide 2189 (T–G) resulting in amino acid substitutions. The corresponding wild-type sequence is shown for comparison.
1339
EGFR phosphorylation analysis Immunohistochemical analyses were performed on 45 tumors including 30 tumors with EGFR mutation analysis. The results of a representative case are shown in Figure 3. The activation EGFR is reflected by its tyrosine-phosphorylating status. The p-EGFR status was correlated with the response to gefitinib. The odds of the respone rate were higher for patients with positive p-EGFR than for those with negative p-EGFR (OR Table 7. Factors associated with gene mutations (n = 30) Mutation
Non-mutation
Male
3/16
13/16
Female
9/14
5/14
Adenocarcinoma
11/21
10/21
Nonadenocarcinoma
1/9
8/9
Non-smokers
11/18
7/18
Former smokers
1/12
11/12
Responders
8/9
1/9
Non-responders
4/21
17/21
Survival functions
1.2
0 = non, 1=mutation
1.0
1.00 1.00-censored
Cum survival
0.8
0.00
0.6
0.00-censored
0.4 0.2 0.0 0.2 0
2
4
6
8
10
12
14
PFS Survival functions 1.2
0 = non, 1=mutation 1.00
Cum survival
1.0
1.00-censored 0.00
0.8
0.00-censored
0.6 0.4 0.2 0
0
10
20
30
OS Figure 2. Survival curve for patients with or without mutation. (A) Kaplan–Meier plots showing progress-free survival (PFS) according to EGFR mutation (n = 30). (B) Kaplan–Meier plots showing overall survival (OS) according to EGFR mutation (n = 30).
2.071; 95% CI 0.955–4.491; P = 0.013). There was a tendency for prolonged PFS and OS, but this was not statistically significant. The RR, PFS and OS for patients with positive and negative p-EGFR are shown in Table 8. Six of 12 tumors with the EGFR mutation, but only four of the 18 tumors without this mutation, were p-EGFR positive.
Discussion
x
P-value
6.451
0.014
4.471
0.040
8.356
0.005
12.804
0.001
The outcome for patients with NSCLC who fail to respond or have progressive disease after first-line chemotherapy is poor. Docetaxel is considered the standard second-line chemotherapy for NSCLC. Alimtaw has recently been approved by the US Food and Drug Administration (FDA) for patients with NSCLC in whom platinum-based chemotherapy has failed. However, the therapeutic efficacy of these drugs is limited [16, 17]. Erlotinib has shown survival benefit for patients who previously failed at least one prior chemotherapy [18]. In two large phase II trials (IDEAL1 and IDEAL 2), following platinum-containing therapies, gefitinib still has
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Mutations of EGFR were detected in 12 tumors. The clinical features and the results of mutational study are listed in Table 6. Four were in-frame deletions in exon 19 (delE746A750, delL747-P753 insS). Eight mutations were in exon 21. All were the 2819T ! G, which resulted in an L858 change. Representative cases are shown in Figure 1. Eight of the nine patients who responded to gefitinib had the EGFR gene mutation in contrast with none of the nine patients with disease progression. Four of the 12 patients with disease control had the mutation. All except one of these patients had adenocarcinoma, and only one was a former smoker. The difference between the mutation rates in responders and non-responders is statistically significant (P = 0.001). Mutation rates were higher among non-smokers (11/18) than among former smokers (1/12) (P = 0.005). Eleven of 21 patients with adenocarcinoma, but only one of nine patients with non-adenocarcinoma, had a gene mutation (P = 0.040). The mutation rate was also higher among female patients (9/14) than among male patients (3/16) (P = 0.014). The factors associated with the mutation are shown in Table 7. The median PFS for patients with the mutation was 10.0 ± 1.2 months (95% CI 7.6–12.4), whereas it was 3.0 ± 1.1 months (95% CI 0.9 –5.2) for patients without the mutation. The median OS for patients without the mutation was 7.0 ± 2.0 months (95% CI 3.1 –10.9), but has not yet been reached for patients with the mutation. Nine of 12 patients are still alive with a median follow-up of 17.5 (4.6 – 25.3) months. The PFS and OS for patients with the mutation were significantly longer than for patients without the mutation (P-values of 0.0454 and 0.0022, respectively). The PFS and OS for patients with and without the mutation are shown in Figure 2.
1340
an objective remission rate of 12 –18% with a disease control rate of about 50%. On the basis of these two studies, gefitinib has been approved for the treatment of patients with NSCLC who have failed to respond to other chemotherapies [19]. In the present study, the response rate was 31.6% and the disease control rate was 67.3%, both of which are higher than in the phase II trials described above but similar to results obtained in Japan. A total of 102 Japanese patients with advanced lung cancer were evaluated in IDEAL1. Similar to our result, the RR for Japanese patients was higher than that of non-Japanese patients (27.5% compared with 10.45%; OR 3.72; P = 0023). However, the difference in RR was imputed
Table 8. Response rate, progress-free survival and overall survival among patients with positive p-EGFR and negative p-EGFR Positive p-EGFR
Negative p-EGFR
x/log rank
P-value
Responders
6/10
4/10
7.305
0.013
Non-responders
6/35
29/35
PFS [M ± SE (95% CI)]
9.80 ± 2.07 (5.75 –13.85)
5.70 ± 1.87 (2.04–9.36)
3.06
0.0805
OS [M ± SE (95% CI)]
18.70 ± 2.32 (14.16–23.24)
9.80 ± 2.17 (5.54–14.06)
3.76
0.0526
RR, response rate; PFS, progress free survival; OS, overall survival; p-EGFR, phosphorylation epidermal growth factor receptor.
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
Figure 3. Immunohistochemical staining with p-EGFR antibody (magnification 200).
to bias of the baseline predictive factors of the patients (i.e. gender, PS and pathology) rather than ethnicity [9, 20]. In our study, logistic regression analysis showed that histology, gender and smoking history were prognostic factors associated with an objective tumor response. Physical status, prechemotherapy and TNM stage were not predictive factors for tumor response. The data were similar to the two phase II trials. The present study demonstrated that a daily oral dose of 250 mg of gefitinib was well tolerated in Chinese patients. Similar to prior clinical trials evaluating the toxicity of gefitinib, the most frequently reported adverse events were acne-like rash and grade 1 and 2 diarrhea. Severe adverse events were rare. It has been reported that 291 (1.7%) of 17 500 patients treated with ZD1839 in Japan had developed suspected interstitial pneumonia or acute lung injury in contrast with a lower incidence worldwide [21]. We paid more attention to this severe adverse event in our study because of the similarity in ethnicity between Chinese and Japanese, but we did not observe any evidence of interstitial pneumonia. The most recent study showed that EGFR gene mutations were correlated with response to gefitinib in NSCLC. All these mutations were within exons 18 –21 of the kinase domain of EGFR. The mutations were most frequently found in patients who responded to gefitinib, patients with adenocarcinoma, patients who were non-smokers and female patients [12, 13]. In the present study, a total of 12 mutations were detected, with four in-frame deletion mutations in exon 19 and eight missense mutations in exon 21. The mutation rate was significantly higher for gefitinib responders, non-smokers, female patients and patients with adenocarcinoma than for gefitinib non-responders, male patients, former smokers and patients with non-adenocarcinoma. These data agreed with the above studies and clinical data. The previous study showed that the gene mutation rate of Japanese (26%) was much higher than that of non-Japanese. The mutation rate of Japanese patients with adenocarcinoma was 46% [12]. That may partially explain the different response rates of Japanese and Caucasians. The therapeutic efficacy for Chinese patients in the present study was close to that for Japanese which suggests that the EGFR gene mutation rate would be higher in the Chinese population because of the ethnic similarity. EGFR is frequently expressed in NSCLC, and EGFR activation after ligand binding is believed to be its most important
1341
Conclusion In conclusion, a daily oral dose of gefitinib 250 mg is well tolerated in Chinese patients with advanced NSCLC and provides significant antitumor activity in patients who have failed to respond to previous platinum-based chemotherapy. The gene mutation rate of EGFR kinase domain was much higher among gefitinib responders than among non-responders.
Patients with mutation have a longer OS than patients without mutation. Therefore gefitinib can be used as second- or thirdline therapy for treatment of Chinese patients with advanced NSCLC, and gene mutation of the EGFR kinase domain can be used as a predictive marker for patient selection.
References 1. Lung Cancer Collaborative Group. Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised clinical trails. BMJ 1995; 311: 899–909. 2. Schill J. Current standards of care in small-cell and non-small-cell lung cancer. Oncology 2001; 61: 3– 13. 3. Haura EB. Treatment of advanced non-small-cell lung cancer: a review of current randomized clinical trials and an examination of emerging therapies. Cancer Control 2001; 8: 326–336. 4. Moscatello DK, Holgado-Madruga M, Godwin AK et al. Frequent expression of a mutant epidermal growth factor receptor in multiple human tumors. Cancer Res 1995; 55: 5536–5539. 5. Woodburn JR. The epidermal growth factor receptor and its inhibition in cancer therapy. Pharmacol Ther 1999; 82: 241 –250. 6. Baselga J. Why the epidermal growth factor receptor? The rationale for cancer therapy. Oncologist 2002; 7 (Suppl 4): 2–8. 7. Ranson M, Hammond LA, Ferry D et al. ZD1839, a selective oral epidermal growth factor receptor-tyrosine kinase inhibitor, is well tolerated and active in patients with solid, malignant tumors: results of a phase I trial. J. Clin Oncol 2002; 20: 2240–2250. 8. Baselga J, Rischin D, Ranson M et al. Phase I safety, pharmacokinetic, and pharmacodynamic trial of ZD1839, a selective oral epidermal growth factor receptor tyrosine kinase inhibitor, in patients with five selected solid tumor types. J Clin Oncol 2002; 20: 4292–4302. 9. Fukuoka M, Yano S, Giaccone G. Multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small-cell lung cancer. J Clin Oncol 2003; 21: 2237–2246. 10. Kris MG, Natale RB, Herst RS et al. Efficacy of gefitinib, an inhibitor of the epidermal growth factor tyrosine kinase, in symptomatic patients with platinum-based chemotherapy. J Clin Oncol 2000; 18: 2095–2103. 11. Bailey R, Kris M, Wolf M et al. Gefitinib (‘Iressa’, ZD 1839) monotherapy for pretreated advanced non-small cell lung cancer in IDEAL 1 and 2: tumor response is not clinically relevantly predictable from tumor EGFR membrane staining alone. Lung Cancer 2003; 41 (Suppl 571): (Abstr). 12. Paez JG, Janne PA, Lee JC et al. EGFR mutation in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004; 304: 1497–1500. 13. Lynch TJ, Bell DW, Sordella R et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of nonsmall-cell lung cancer to gefitinib. N Engl J Med 2004; 350: 2129–2139. 14. Response evaluation criteria in solid tumors (RECIST) quick reference. Available online at: http://www.nci.nih.gov/bip/RECIST.htm. 15. Cancer Therapy Evaluation Program, Common Toxicity Criteria, Version 2.0, DCTD, NCI, NIH, DHHS. Available online at: http://ctep. cancer.gov. 16. Shepheral FA, Dancey J, Ramlan R et al. Prospective randomized trial of docetaxel versus best supportive care in patients with non-small-cell lung cancer previously treated with platinum-based chemotherapy. J Clin Oncol 2000; 18: 2095–2103.
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
role in clinical settings. The relationships of EGFR expression and EGFR activation status to the sensitivity of EGFR inhibitors have been studied in a few retrospective series. The predictive association of EGFR and p-EGFR with tumor sensitivity has not been verified, but the activation of the downstream signaling molecule (p-Akt) seems to be highly sensitive to gefitinib in NSCLC [22–25]. In the present study, patients with positive p-EGFR expression had a higher response rate (P = 0.013) and a tendency to have longer median PFS and OS (P-values of 0.0805 and 0.0526, respectively). The relationships between p-EGFR activation and activation of the downstream signaling pathways, as well as the relationship of activation status about EGFR and downstream signaling pathways to the response of gefitinib need further investigation. In the present study, EGFR phosphorylation was detected by immunohistochemical analysis in six out of 12 tumors with EGFR mutation in contrast with four out of 19 tumors without EGFR mutation. The data agreed with the results obtained by Huang et al. [26]. In their study, increased EGFR phosphorylation was detected by immunoblot in 26 out of 39 tumors with EGFR mutation compared with 14 out of 38 tumors without mutation. In general, the EGFR kinase domain mutations seem to have a different effect on kinase activation. The functional significance of mutation on phosphorylation requires additional investigation. Although previous studies have shown that the EGFR kinase mutations are correlated with clinical response, it is possible that not all patients with EGFR mutation respond to gefitinib. In the present study, four patients with disease control also had EGFR gene mutation, which has not been documented in other studies. All except one of these four patients had histological evidence of adenocarcinoma and OS was >10 months (follow-up 10– 25 months). Our data suggest that even patients with gene mutation who failed to show objective tumor response could probably also benefit from gefitinib treatment. In general, OS for patients with mutation was significant longer than that for patients without mutation. Further study of the antitumor mechanism of gefitinib may provide better selection criteria for patients. We should note that all mutations were performed with tissues at diagnosis. Could this gene status represent that after exposure to gefitinib? Does cytotoxic therapy affect the mutation status? At present, we lack information on the incidence of mutation after cytotoxic therapy. Further investigations are needed to determine the true incidence in untreated and treated tumors.
1342
23.
24.
25.
26.
the 94th Annual Meeting of the American Association for Cancer Research. Washington, DC; July 2003. Philadelphia, PA: AACR, 1362. Cappuzzo F, Gregorc V, Rossi E et al. Gefitinib in pretreated nonsmall cell lung cancer (NSCLC): Analysis of efficacy and correlation with HER2 and epidermal growth factor receptor expression in locally advanced or metastatic NSCLC. J Clin Oncol 2003; 21: 2658–2663. Han S-W, Hwang PG, Chung DH et al. Epidermal growth factor receptor (EGFR) downstream molecules as response predictive markers for gefitinib (Iressa, ZD 1839) in chemotherapy-resistant non small cell lung cancer. Int J Cancer 2005; 113: 109– 115. Cappuzzo F, Magrini E, Cereoli GL et al. Akt phosphorylation and gefitinib effficacy in patients with advanced non-small cell lung cancer. Int J Cancer 2004; 96: 1133–1141. Huang SF, Liu HP, Li LH et al. High frequency of epidermal growth factor receptor mutations with complex patterns in non-small cell lung cancer related to gefitinib responsiveness in Taiwan. Clin Cancer Research 2004; 10: 8195–8233.
Downloaded from http://annonc.oxfordjournals.org/ at University of Saskatchewan Library on November 28, 2012
17. De Marinis F, De Petris L. Pemetrexed in second-line treatment of non-small-cell lung cancer. Oncology (Huntingt) 2004; 18 (Suppl 8): 38 –42. 18. New drug to treat most common lung cancer. FDA Consum 2005; 39: 3. 19. Cohen MH, Williams GA, Sridhara R et al. FDA drug approval summary: gefitinib (ZD1839) (Iressa) tablets. Oncologist 2003; 8: 303 –306. 20. Nakagawa K, Tamura T, Negoro S et al. Phase I pharmacokinetic trial of the selective oral epidermal growth factor receptor tyrosine kinase inhibitor gefitinib (‘Iressa’, ZD1839) in Japanese patients with solid malignant tumors. Ann Oncol 2003; 14: 922–930. 21. Inoue A, Saijo Y, Maemondo M, Gomi K et al. Severe acute interstitial pneumonia and gefitinib. Lancet 2003; 361: 137–139. 22. Bailey LR, Kris M, Wolf M et al. Tumor EGFR membrane staining is not clinically relevant for predicting response in patients receiving gifitinib (Iressa, ZD 1839) monotherapy for pretreated advanced non-smallcell lung cancer: IDEAL 1 and 2 (abstract no. LB-170). Proceedings of