Combination Therapy With Gefitinib and Rofecoxib in ...

VOLUME

25

䡠

NUMBER

22

䡠

AUGUST

1

2007

JOURNAL OF CLINICAL ONCOLOGY

O R I G I N A L

R E P O R T

Combination Therapy With Gefitinib and Rofecoxib in Patients With Platinum-Pretreated Relapsed Non–Small-Cell Lung Cancer Kenneth J. O’Byrne, Sarah Danson, David Dunlop, Nick Botwood, Fumiko Taguchi, David Carbone, and Malcolm Ranson From the St James’s Hospital, Dublin, Ireland; Christie Hospital, Manchester; Beatson Oncology Centre, Glasgow, Scotland; AstraZeneca, Macclesfield, United Kingdom; and Vanderbilt-Ingram Cancer Center, Nashville, TN. Submitted October 26, 2006; accepted April 24, 2007. Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article. Address reprint requests to Kenneth J. O’Byrne, MD, St James’s Hospital, Dublin, Ireland; e-mail: kobyrne@ stjames.ie. © 2007 by American Society of Clinical Oncology 0732-183X/07/2522-3266/$20.00 DOI: 10.1200/JCO.2006.09.2791

A

B

S

T

R

A

C

T

Purpose In non–small-cell lung cancer (NSCLC), the epidermal growth factor receptor (EGFR) and cyclooxygenase-2 (COX-2) play major roles in tumorigenesis. This phase I/II study evaluated combined therapy with the EGFR tyrosine kinase inhibitor (TKI) gefitinib and the COX-2 inhibitor rofecoxib in platinum-pretreated, relapsed, metastatic NSCLC (n ⫽ 45). Patients and Methods Gefitinib 250 mg/d was combined with rofecoxib (dose escalated from 12.5 to 25 to 50 mg/d through three cohorts, each n ⫽ 6). Because the rofecoxib maximum-tolerated dose was not reached, the 50 mg/d cohort was expanded for efficacy evaluation (n ⫽ 33). Results Among the 42 assessable patients, there was one complete response (CR) and two partial responses (PRs) and 12 patients with stable disease (SD); disease control rate was 35.7% (95% CI, 21.6% to 52.0%). Median time to tumor progression was 55 days (95% CI, 47 to 70 days), and median survival was 144 days (95% CI, 103 to 190 days). In a pilot study, matrix-assisted laser desorption/ionization (MALDI) proteomics analysis of baseline serum samples could distinguish patients with an objective response from those with SD or progressive disease (PD), and those with disease control (CR, PR, and SD) from those with PD. The regimen was generally well tolerated, with predictable toxicities including skin rash and diarrhea. Conclusion Gefitinib combined with rofecoxib provided disease control equivalent to that expected with single-agent gefitinib and was generally well tolerated. Baseline serum proteomics may help identify those patients most likely to benefit from EGFR TKIs. J Clin Oncol 25:3266-3273. © 2007 by American Society of Clinical Oncology

INTRODUCTION

Five-year survival rates for patients with non– small-cell lung cancer (NSCLC) are 3% to 7% for stage IIIB, and less than 1% for stage IV disease.1-3 Therapeutic options for relapsed NSCLC are limited. Molecular targeting approaches manipulating the biology of the disease may hold promise for the future. The epidermal growth factor receptor (EGFR) plays a role in tumorigenesis, stimulating cell proliferation, inhibiting apoptotis, and promoting angiogenesis and metastasis.4,5 EGFR overexpression is seen in NSCLC, linked with a poor prognosis, and is a target for anticancer therapy.6 When this study was designed, phase II study data showed that gefitinib, an orally active

EGFR tyrosine kinase inhibitor (TKI), was well tolerated, with antitumor activity.7,8 Cyclooxygenase-2 (COX-2) overexpression in NSCLC9 correlates with poor survival in earlystage disease.10 COX-2 promotes tumor invasion and angiogenesis through induction of prostaglandins and vascular endothelial growth factor (VEGF).11,12 COX-2 inhibitors (COX-2Is) inhibit the proliferation and invasiveness of NSCLC in vitro and enhance the cytotoxicity of chemotherapeutic agents.6,13 Selective COX-2Is are nonsteroidal anti-inflammatory drugs (NSAIDs). Phase II studies suggest that COX-2Is may enhance neoadjuvant and first-line chemotherapy in early-stage14 and advanced NSCLC15; however, this effect was not seen with second-line docetaxel therapy.16,17

3266

Downloaded from jco.ascopubs.org on November 4, 2015. For personal use only. No other uses without permission. Copyright © 2007 American Society of Clinical Oncology. All rights reserved.

Gefitinib Plus Rofecoxib in Pretreated NSCLC

Although EGFR activation may induce COX-2 expression,18,19 no relationship is seen between EGFR and COX-2 expression in NSCLC tumors.9 EGFR-TKIs and COX-2Is block different cancer signaling pathways. Therefore, their combination may be a useful anticancer strategy.4,9 That gefitinib and the COX-2I celecoxib additively or synergistically inhibit head and neck squamous cell carcinoma growth in vitro, inducing G(1) arrest and apoptosis and suppression of endothelial capillary formation,20 supports this contention and is consistent with other literature.21,22 Collectively, these data provided a rationale for combining a COX-2I with gefitinib in NSCLC. The safety and efficacy of gefitinib 250 mg/d in combination with the COX-2I rofecoxib (at three dose levels: 12.5, 25, or 50 mg/d) in advanced, pretreated NSCLC was investigated. The value

of baseline serum proteomics analysis in predicting response to therapy was evaluated. PATIENTS AND METHODS Study Design This was a three-center, single-arm, open-label, noncomparative trial. Patients (aged ⱖ 18 years) with histologically/cytologically confirmed inoperable stage III or IV NSCLC and platinum-pretreated, relapsed disease were eligible. At least one measurable lesion according to Response Evaluation Criteria in Solid Tumors (RECIST) and a WHO performance status of 2 or lower was required. Exclusion criteria included coexisting malignancies diagnosed within the last 5 years (except basal cell carcinoma or cervical cancer in situ); incomplete

Table 1. Patient Demographic and Tumor Response Data Rofecoxib Dose Level Characteristic Age, years Median Range Sex, No. Male Female WHO performance status, No. 0 1 2 Disease stage, No. IIIa IIIb IV Histology, No. Adenocarcinoma Bronchoalveolar carcinoma Squamous-cell carcinoma Undifferentiated carcinoma Metastases, No.ⴱ Adrenal Liver Bone Lymph nodes Lung Skin/soft tissue Other Prior chemotherapies, No.† 1 2 Smoking status, No.‡ Current Ex-smoker Never-smoker Best response, No.§ Complete/partial Stable disease Progressive disease

Dose Level 1 12.5 mg/d (n ⫽ 6)

Dose Level 2 25 mg/d (n ⫽ 6)

Dose Level 3 50 mg/d (n ⫽ 33)

All (N ⫽ 45)

65 53-69

62 54-80

61 44-76

62 44-80

4 2

3 3

22 11

29 16

1 4 1

0 4 2

5 22 6

6 30 9

0 1 5

0 0 6

2 4 27

2 5 38

2 0 1 3 4 0 1 0 1 4 0 0

2 0 4 0 5 0 2 0 1 4 0 0

15 2 12 4 30 4 7 3 8 22 3 7

19 2 17 7 39 4 10 3 10 30 3 7

4 2

5 1

22 11

31 14

2 3 1

2 3 1

13 15 4

17 21 6

1 1 4

2 1 3

0 10 20

3 12 27

ⴱ

Patients could have more than one metastatic site. †One patient did not receive prior platinum chemotherapy and was subsequently withdrawn from the study. ‡Smoking history available for 44 patients. §Forty-two patients were assessable for response.

3267

www.jco.org


O’Byrne et al

healing from previous surgery; unresolved chronic toxicity greater than National Cancer Institute Common Toxicity Criteria (NCI-CTC) version 2.0 grade 2 from prior anticancer therapy; clinically active interstitial lung disease; severe or uncontrolled systemic disease; treatment with chemotherapy, hormone therapy, or immunotherapy within 21 days before enrollment; and pregnancy or current breastfeeding. The primary objective was to evaluate the tolerability of gefitinib 250 mg/d in combination with rofecoxib 12.5, 25, or 50 mg/d. Secondary objectives were the efficacy of the combination in terms of objective response (complete response [CR] or partial response [PR]) and disease control (CR, PR, and stable disease [SD]) rates, time to tumor progression and overall survival, the pharmacokinetic effect of rofecoxib on gefitinib, and the role of serum proteomics in identifying patients benefiting from treatment. The trial was performed in accordance with the Declaration of Helsinki, Good Clinical Practice guidelines, and appropriate regulatory requirements. Local ethics committee approval was obtained, and all patients provided written, informed consent. The study consisted of two phases: dose-finding and efficacy. Eligible patients entered the dose-finding phase of the study and received gefitinib 250 mg/d for 7 days. On day 8, rofecoxib was added to gefitinib 250 mg/d in three cohorts at three different dose levels: 12.5 mg/d (dose level 1), 25 mg/d (dose level 2), and 50 mg/d (dose level 3). Combination treatment continued until day 28, when tolerability assessments were performed. The maximumtolerated dose (MTD) of rofecoxib was defined as the dose immediately preceding that at which more than one in six patients experienced grade 3/4 nonhematologic toxicity (dose-limiting toxicity [DLT]). If more than one in six patients in a dose level experienced DLT, either, in the case of dose level 1, the trial was to be stopped with no expansion of an MTD cohort or, in the cases of dose levels 2 and 3, the rofecoxib dose was reduced to the previous dose level. However, if toxicity was grade 2 or lower, patients entered the next cohort. In the subsequent efficacy phase, patients received gefitinib 250 mg/d and rofecoxib at its MTD (up to 50 mg/d) until disease progression or unacceptable toxicity. Tolerability Adverse events (AEs) and their causality and severity using NCI-CTC criteria were recorded. Grade 3/4 toxicity was managed by dose interruption (up to 14 days) of gefitinib and/or rofecoxib. Repeat dose interruptions were allowed to manage toxicity. If management was insufficient for gefitinibrelated toxicity, patients were withdrawn. If management was insufficient for rofecoxib, dose reduction to the lower dose level was permitted. Any patient receiving rofecoxib 12.5 mg/d who experienced unacceptable toxicity had rofecoxib discontinued but continued gefitinib if gaining clinical benefit. Efficacy Patients underwent computer tomography scans every 2 months to determine RECIST response. CR was defined as the disappearance of all target lesions, PR at least a 30% decrease in the sum of the longest diameter of target lesions, progressive disease (PD) at least a 20% increase in the sum of the longest diameter of target lesions, and SD neither sufficient shrinkage to qualify for PR nor increase to qualify for PD. Objective responses (CR or PR) were confirmed 4 or more weeks after the criteria for response were first met. Disease control rate was defined as CR plus PR plus SD. Progression-free survival was determined from date of enrollment to date of disease progression (PD) or death. Overall survival was determined from date of enrollment until death or the last date the patient was recorded alive.

tography with tandem mass spectrometry at a central laboratory (AstraZeneca, Alderley Park, United Kingdom). Matrix-Assisted Laser Desorption/Ionization–Proteomic Profiling of Serum Samples Serum samples obtained from patients before gefitinib and rofecoxib treatment were stored at ⫺80°C until analysis. Each sample was thawed on ice and diluted to 10% in H2O, and 1 ␮L was mixed with the same amount of matrix (50% acetonitrile, 0.1% trifluoroacetic acid, 35 mg/mL sinnapinic acid) on a gold plate. Protein expression profiles were analyzed with a Voyager STR matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometer (PerSeptive Biosystems, Framingham, MA). Each spectrum was the average of two spectra (each the result of 250 laser shots randomly acquired over a serum spot). Spectra were internally calibrated with peaks of hemoglobin-␤ and apolipoprotein C-1 (APOC-1). The baseline of each spectrum was corrected using Data Explorer software (Applied Biosystems, Foster City, CA). Signals in 2,000 to 20,000 M/Z were considered, binned, and normalized as previously reported.23 Statistical Analysis A cohort size of six patients was considered sufficient to detect DLTs during the dose-finding phase of the study. During the efficacy phase, a sample size of 33 patients was considered sufficient to give at least 80% probability of rejecting a baseline response rate of 5% with an exact 5% one-sided significance test when the true response is at the clinically relevant rate of 20%. The intention-to-treat (ITT) population, comprising all patients enrolled who received medication, was used to analyze efficacy parameters. Tumor response rate was summarized by proportion and 90% CIs. Progression-free and overall survival were summarized using the KaplanMeier method, along with the appropriate 95% CIs. The weighted flexible compound covariate method (WFCCM)24 was used to verify whether the proteomic patterns could be used to classify serum samples into two classes according to response to treatment as follows: (a) [CR and PR] versus [SD and PD] or (b) [CR, PR, and SD] versus PD. The agglomerative hierarchical clustering algorithm was also applied to define the pattern among the significant discriminators in this cohort.

RESULTS

Patients A total of 45 patients were recruited to the study (Table 1), with the first patient enrolled in June 2002. Adenocarcinoma and

Patients (%) 0 Rash† ‡

Diarrhea Dry skin Pruritus

3268

Stomatitis Dyspepsia

30

40

50

60

70

80

90

100

G3/4* G1/2 G3/4* G1/2 G3/4 G1/2 G3/4 G1/2 G3/4 G1/2 G3/4 G1/2

G3/4 Abdominal G1/2

Pain§

20

G1/2

Nausea

Pharmacokinetics Blood samples (5 mL) were taken (a) at predose on day 7 and after 1, 3, 7, and 24 hours after administration of gefitinib 250 mg/d, and (b) predose on day 28 and after 1, 3, 7, and 24 hours after administration of gefitinib and rofecoxib combination therapy to evaluate the pharmacokinetic effect of rofecoxib by comparison of the area under the plasma concentration-time curve during the dosing interval (AUC0-␶) and minimum plasma concentration (Cmin) of gefitinib alone (day 7) and when given with rofecoxib (day 28). After immediate centrifugation (10 minutes at 1,000 ⫻ g), plasma samples were stored at ⫺20°C. Samples were analyzed by high-performance liquid chroma-

10

Gefitinib-related Rofecoxib-related

G3/4

Fig 1. Treatment-related adverse events occurring in at least 5% of patients and categorized by causality. ⴱGrade 3 only. †Includes rash follicular; rash not otherwise specified (NOS); dermatitis acneiform; rash pustular. ‡Includes diarrhea NOS; loose stools. §Includes abdominal pain NOS; abdominal pain upper. G, grade. JOURNAL OF CLINICAL ONCOLOGY



A

Pre

B

Post

Fig 2. The (A) pre- and (B) post-treatment chest x-rays and soft tissue computed tomography images of the patient whose response was difficult to measure by Response Evaluation Criteria in Solid Tumors criteria are shown. There is clearly a greater than 50% resolution of the tumor mass. In all, of the 42 patients with measurable disease at baseline, there were three patients with objective tumor responses, 12 with stable disease, and 27 with disease progression.

squamous cell carcinoma (SCC) were the most common primary tumors, accounting for 42% and 38% of patients, respectively. Of these, 21 had received carboplatin and 23 cisplatin regimens combined with vinblastin/mitomycin (n ⫽ 11), ifosfamide/mitomycin (n ⫽ 7), vinorelbine (n ⫽ 3), docetaxel (n ⫽ 12), or gemcitabine (n ⫽ 11). Fourteen patients received second-line chemotherapy, eight with docetaxel. One patient recruited had not received platinumbased therapy and was subsequently withdrawn from the study. Treatment Patients were entered into all cohorts: dose level 1 (n ⫽ 6), dose level 2 (n ⫽ 6), and dose level 3 (n ⫽ 33). All received at least one dose of gefitinib-rofecoxib combination therapy and were included in the ITT population. Overall, patients received treatment with gefitinib for a median of 56 days (range, 6 to 487 days). MTD During the dose-finding stage, no patient in any of the cohorts experienced grade 3/4 nonhematologic toxicity; therefore, the MTD for rofecoxib was not defined. For this study, dose level 3 (rofecoxib

50 mg/d) was assumed to be the MTD and subsequently expanded to 33 patients for safety and efficacy assessment. Tolerability Thirty-one (69%) of 45 patients experienced treatmentrelated AEs, most of which were mild/moderate. Grade 1/2 rash and diarrhea in 14 (31.1%) and 14 (31.1%) patients, respectively, were the most common gefitinib-related AEs (Fig 1). Rofecoxibrelated AEs included grade 1/2 dyspepsia (6.7%) and abdominal pain (6.7%). Five patients experienced gefitinib-related grade 3/4 AEs: grade 4 exacerbation of dyspnea (n ⫽ 1 in dose level 3 [rofecoxib 50 mg/d]), grade 3 diarrhea (n ⫽ 2; both in dose level 3 [rofecoxib 50 mg/d]), grade 3 rash (n ⫽ 2; one patient in dose level 2 [rofecoxib 25 mg/d], and one patient in dose level 3 [rofecoxib 50 mg/d]). No grade 3/4 toxicity was attributed to rofecoxib, or combined rofecoxib and gefitinib. There were no drug-related toxic deaths. One patient in dose level 3 (rofecoxib 50 mg/d) died as a result of a cerebrovascular accident that was not considered to be drug related on day 28 of treatment. 3269

www.jco.org


In eight patients, treatment-related AEs led to treatment interruptions: one in dose level 2 (gefitinib); seven in dose level 3 (gefitinib in two patients, rofecoxib in one, and both drugs in four). One patient in dose level 1 experienced grade 1 renal impairment attributed to rofecoxib; rofecoxib was withdrawn and gefitinib continued. One patient experienced grade 4 dyspnea attributed to gefitinib, which was withdrawn. Efficacy Of 42 patients assessable according to RECIST criteria, two (4.8%; 95% CI, 0.6% to 16.2%) responded to treatment (one male, one female) and 13 patients had SD (11 male, two female), giving a 35.7% disease control rate (95% CI, 21.6% to 52.0%). The male patient with the CR (duration of response, 197 days) had SCC and a 48 pack-year history of cigarette consumption, and also smoked cigars and a pipe. The female patient with the PR (duration of response, 190 days) had adenocarcinoma and was an ex-smoker. A third patient, a nonsmoker with adenocarcinoma, was considered by the investigator to have a PR not measurable by RECIST criteria because of the diffuse nature of the response (Fig 2). Overall, smoking history was available for 41 of these patients. Four of 16 current smokers (25%; one CR; three SD), nine of 20 ex-smokers (45%; one PR, eight SD), and two of five never-smokers (40%; one PR, one SD) had disease control for at least 4 months. In the ITT population, median time to progression was 55 days (95% CI, 47 to 70 days; Figure 3A), and median overall survival was

A Proportion Event Free

1.00

Median time to progression: 55 days (95% CI, 47 to 70) 6-month progression-free survival rate: 17.8% (95% CI, 6.6 to 28.9)

0.75

0.50

0.25

Mean Gefitinib (AUC0-τ ng•h/mL)

O’Byrne et al

Gefitinib alone 12.5 mg rofecoxib 25 mg rofecoxib 50 mg rofecoxib

15,000

10,000

5,000

0

Dose level 1

Dose level 2

Dose level 3

(n = 5) (n = 6)

(n = 6) (n = 6)

(n = 21) (n = 16)

Cohort Fig 4. The impact of rofecoxib on gefitinib pharmacokinetics 24 hours posttreatment in each patient cohort was studied. Mean area under the plasma concentration-time curve during the dosing interval (AUC0-␶) did not differ significantly for gefitinib alone compared with gefitinib in combination with rofecoxib 12.5, 25, or 50 mg.

144 days (95% CI, 103 to 190 days). The proportion of patients alive at 6 months was 40.3% (95% CI, 25.6% to 55.0%; Figure 3B). Pharmacokinetics Blood samples from 33 patients were studied. Mean AUC0-␶ did not differ markedly for gefitinib alone compared with gefitinib in combination with rofecoxib 12.5, 25, or 50 mg (Fig 4). Proteomics Baseline serum samples from 34 patients were analyzed. Statistical analysis demonstrated that 55 mass spectroscopy signals were differentially expressed between responders (n ⫽ 3) and SD/PD (n ⫽ 31; Fig 5A), and 90 between the disease control group (CR, PR, and SD patients; n ⫽ 14) and the PD group (n ⫽ 20; Fig 5B). Using WFCCM based on these signals, all 34 patients were correctly classified. The agglomerative hierarchical clustering algorithm showed that the selected proteomics pattern distinguished response patterns with just one misclassification between the disease control group and the PD group.

Censored n = 45

0

50

100

150

200

250

300

350

400

450

500

DISCUSSION

Time to Progression (days)

B Proportion Event Free

1.00

Median survival time: 144 days (95% CI, 103 to 190) 6-month overall survival rate: 40.3% (95% CI, 25.6 to 55.0)

0.75

0.50

0.25 Censored n = 45

0

100

200

300

400

500

600

Overall Survival Time (days) Fig 3. (A) Time to disease progression and (B) overall survival in the intentionto-treat population. Circles represent censored observations. 3270

The combination of gefitinib 250 mg/d and the COX-2I rofecoxib up to 50 mg/d is well tolerated in advanced, platinum-pretreated NSCLC. The incidence of gefitinib-related AEs such as mild to moderate rash and diarrhea was similar to that reported in the ISEL (IRESSA Survival Evaluation in Lung cancer) trial.25 Minor rofecoxib-associated GI AEs were similar to those seen in patients with osteoarthritis and rheumatoid arthritis.26,27 Since this study was completed, rofecoxib has been withdrawn from clinical use. In the APPROVe (Adenomatous Polyp Prevention On Vioxx) study, long-term use (⬎ 18 months) of rofecoxib in patients with intestinal polyps and no history of cardiovascular disease showed an increased risk of confirmed serious thrombotic events (including myocardial infarction and stroke) compared with placebo.28,29 US and European regulatory authorities now require the prescribing information of the COX-2I class to include safety restrictions concerning the increased risk of vascular AEs.30 One patient in our JOURNAL OF CLINICAL ONCOLOGY



PR---s3 SD--s10 PR---s2 SD---s8 SD---s6 SD---s7 SD--s11 SD--s12 CR---s1 SD---s5 SD--s13 SD---s4 SD--s14 SD---s9 PD-----s21 PD-----s15 PD-----s28 PD-----s16 PD-----s33 PD-----s17 PD-----s23 PD-----s19 PD-----s24 PD-----s27 PD-----s18 PD-----s25 PD-----s29 PD-----s20 PD-----s32 PD-----s26 PD-----s22 PD-----s30 PD-----s31 PD-----s34

B

CR-------s1 PR-------s2 PR-------s3 SD---s6 SD--s11 SD--s12 SD---s5 SD--s10 PD----s26 PD----s28 PD----s31 SD---s7 PD----s20 SD---s9 SD---s8 PD----s16 SD---s4 PD----s15 SD--s13 PD----s34 PD----s30 PD----s19 PD----s24 PD----s21 PD----s29 SD--s14 PD----s18 PD----s25 PD----s27 PD----s17 PD----s22 PD----s23 PD----s33 PD----s32

A

2736.224854 4532.075195 4750.34375 2202.638428 5044.393555 6567.771484 9468.350586 3078.319824 3285.180664 16117.28809 2396.950684 2565.084961 3116.12793 3162.589111 3062.194824 4916.527344 5321.429199 2428.081055 6140.060547 2910.775146 3874.096436 3257.504395 15925.625 2511.943115 8043.841309 4601.681152 6003.761719 3074.546631 4923.147949 3837.374512 10954.44922 2905.839355 4292.939453 3754.09668 5806.587891 9437.052734 2337.453857 2903.407715 2142.493164 4150.706543 5796.099121 2543.700928 3961.759521 4176.397949 4439.831543 5262.373047 5782.408203 2814.644775 3105.053955 3889.293701 3415.381348 2080.189453 7929.51709 4227.44873 3850.917236 5581.364746 8122.714844 10560.41016 10593.01563 7257.53418 8289.513672 3001.088867 5399.090332 2732.384766 7552.561523 11857.99219 6584.750488 13306.96191 11140.1709 3050.108643 3167.711426 3635.920898 4233.505859 2682.624756 8352.740234 14637.55469 7527.102051 14886.4541 7589 14605.11035 6403.311523 7285.727539 2975.246094 2316.980469 5821.276855 3110.231201 9594.073242 5727.086914 4860.934082 10275.29004

3078.319824 2064.983398 10560.41016 6042.05957 5218.958496 5240.240723 2221.266846 10373.04688 5321.429199 2171.761475 5995.631348 2076.427002 3251.139404 7240.677734 14054.6582 2760.822998 5354.282715 5560.351074 12785.26074 2159.144531 2146.611328 3019.218018 5696.768066 4588.274902 5284.224121 6516.999512 7969.640137 4532.075195 2202.638428 4109.952637 4608.249512 9182.197266 5044.393555 6567.771484 2979.209717 13326.72168 3219.638428 5733.949219 13146.76367 5096.10498 10510.05273 10893.22168 3895.178955 11022.18555 2565.084961 12106.06836 6225.013672 6677.220703 7377.575684 4933.076172 3214.225098 4455.914551 12859.95801 14038.95117 9568.413086

Fig 5. Agglomerative hierarchical clustering of matrix-assisted laser desorption/ionization-time of flight mass spectroscopy signals of (A) responders (n ⫽ 3) versus stable disease (SD)/progressive disease (PD; n ⫽ 31) and (B) disease control (n ⫽ 14) versus PD (n ⫽ 20) is shown with significant differential expression of (a) 55 and (b) 90 signals, respectively.

study experienced an early fatal cerebrovascular accident not considered related to therapy. In the recent ISEL placebo-controlled trial, gefitinib improved time to treatment failure (P ⫽ .0006) and objective response (8.0% v 1.3%; P ⬍ .0001) compared with placebo. Disease control was achieved in 39.7% of patients,25 similar to that seen in the IRESSA Dose Evaluation in Advanced Lung Cancer (IDEAL) randomized phase II studies.7,8 However, the overall survival with gefitinib 250 mg/d monotherapy failed to reach statistical significance (P ⫽ .087).25 In a similar setting, the EGFR-TKI erlotinib improved survival.31 In the present study, 35.7% of patients (12 male and three female patients) achieved disease control, with an objective tumor response seen in three patients (7.1%; one male and two female). These findings suggest that rofecoxib combined with gefitinib is not superior to single-agent gefitinib in platinum-pretreated NSCLC. Pharmacokinetic analyses indicate that rofecoxib (up to 50 mg/d) does not affect exposure to gefitinib (250 mg/d). Therefore, the results cannot be attributed to a detrimental effect of rofecoxib on gefitinib metabolism. Certain clinicopathological features—female sex, nonsmoker status, bronchioalveolar cell carcinoma and adenocarcinomas, and Asian origin—predict an increased likelihood of response to gefitinib therapy.32-34 In keeping with this, two of three responders were female with adenocarcinomas, one a nonsmoker and the other an ex-smoker. In contrast, the patient with a CR had a squamous cell carcinoma and was a current smoker. This observa-

tion, confirmed in other studies, indicates that clinicopathologic parameters alone are insufficient to identify those patients likely to benefit from EGFR-TKIs.25,31 NSCLC tumors with EGFR tyrosine kinase domain mutations are more sensitive to gefitinib.34,35 Fluorescence in situ hybridization– detected increased EGFR gene copy number may also be predictive for a gefitinib treatment effect.36 Unfortunately, in our patients, the EGFR gene mutational or copy number status of the responding tumors is unknown. A number of recent studies have reported EGFR TKI/COX-2I combinations in NSCLC. An interim analysis investigating gefitinib 250 mg/d and the COX-2I celecoxib 400 mg twice daily in 10 platinum-refractory NSCLC patients demonstrated a PR in two patients (20%) and SD in three (30%).37 In a further phase II study, 31 chemotherapy-naı¨ve patients treated with gefitinib and celecoxib 400 mg/d gave a response rate of 16.1%. Median duration of response, progression-free survival, and overall survival were 5.7, 2.8, and 7.2 months, respectively. Finally, a dose-finding study combining celecoxib with erlotinib in 22 patients with advanced NSCLC demonstrated an objective response rate of 33% and disease control rate of 57%. The optimal biologic dose of celecoxib, based on maximal decrease in urinary prostaglandin E-M was celecoxib 600 mg twice daily.38,39 These studies provide encouragement that in selected patients, identified on the basis of clinical and biochemical predictive factors, cotargeting EGFR and COX-2 may be an effective strategy. 3271

www.jco.org


O’Byrne et al

Proteomic analysis, evaluating changes in the expression of a broad range of proteins to a given drug or stimulus, may identify patients likely to benefit from targeted therapies. The results from this study led to subsequent work on pretreatment serum samples from EGFR-TKI–treated patient cohorts. The MALDI-spectra obtained are reproducible within and between institutions. An algorithm developed to predict benefit was validated in subsequent cohorts and is being tested in prospective studies.40,41 In conclusion, rofecoxib combined with gefitinib is well tolerated but does not appear more active than gefitinib alone. Further evaluation of baseline serum proteomics to identify patients likely to benefit from therapy is justified. AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Although all authors completed the disclosure declaration, the following authors or their immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO’s conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

REFERENCES 1. van Zandwijk N, Giaccone G: Treatment of metastatic non-small cell lung cancer. Curr Opin Oncol 8:120-125, 1996 2. Ginsberg RJ, Vokes E, Raben A: Non-small cell lung cancer, in DeVita VT, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology (ed 5). Philadelphia, PA, Lippincott-Raven, 1997, pp 858-911 3. Mountain CF: Revisions in the international system for staging lung cancer. Chest 111:1710-1717, 1997 4. Sporn MB, Todaro GJ: Autocrine secretion and malignant transformation of cells. N Engl J Med 303:878-880, 1980 5. Woodburn JR: The epidermal growth factor receptor and its inhibition in cancer therapy. Pharmacol Ther 82:241-250, 1999 6. Richardson CM, Sharma RA, Cox G, et al: Epidermal growth factor receptors and cyclooxygenase-2 in the pathogenesis of non-small-cell lung cancer: Potential targets for chemoprevention and systemic therapy. Lung Cancer 39:1-13, 2003 7. Fukuoka M, Yano S, Giaccone G, et al: Multiinstitutional randomized phase II trial of gefitinib for previously treated patients with advanced nonsmall-cell lung cancer. J Clin Oncol 21:2237-2246, 2003 8. Kris MG, Natale RB, Herbst RS, et al: Efficacy of gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: A randomized trial. JAMA 290:2149-2158, 2003 9. Richardson CM, Richardson D, Swinson DEB, et al: Cyclooxygenase-2 protein levels are independent of epidermal growth factor receptor expression or activation in operable non-small cell lung cancer. Lung Cancer 48:47-57, 2005 10. Khuri FR, Wu H, Lee JJ, et al: Cyclooxygenase-2 overexpression is a marker of poor prognosis in stage 3272

Employment: Nick Botwood, AstraZeneca Leadership: Nick Botwood, AstraZeneca Consultant: David Carbone, AstraZeneca; Malcolm Ranson, AstraZeneca Stock: Nick Botwood, AstraZeneca Honoraria: David Carbone, AstraZeneca; Malcolm Ranson, AstraZeneca; Kenneth J. O’Byrne, AstraZeneca Research Funds: Kenneth J. O’Byrne, Funds, AstraZeneca Testimony: N/A Other: N/A

AUTHOR CONTRIBUTIONS Conception and design: Kenneth J. O’Byrne, David Dunlop, Nick Botwood, Fumiko Taguchi, David Carbone, Malcolm Ranson Financial support: Kenneth J. O’Byrne, David Dunlop, Nick Botwood, Fumiko Taguchi, David Carbone, Malcolm Ranson Administrative support: Kenneth J. O’Byrne, David Dunlop, Nick Botwood, Fumiko Taguchi, David Carbone, Malcolm Ranson Provision of study materials or patients: Kenneth J. O’Byrne, Sarah Danson, David Dunlop, David Carbone, Malcolm Ranson Collection and assembly of data: Kenneth J. O’Byrne, Sarah Danson, David Dunlop, Fumiko Taguchi, David Carbone, Malcolm Ranson Data analysis and interpretation: Kenneth J. O’Byrne, David Dunlop, Nick Botwood, Fumiko Taguchi, David Carbone, Malcolm Ranson Manuscript writing: Kenneth J. O’Byrne, Sarah Danson, David Dunlop, Nick Botwood, Fumiko Taguchi, David Carbone, Malcolm Ranson Final approval of manuscript: Kenneth J. O’Byrne, Sarah Danson, David Dunlop, Nick Botwood, Fumiko Taguchi, David Carbone, Malcolm Ranson

I non-small cell lung cancer. Clin Cancer Res 7:861867, 2001 11. Tsujii M, Kawano S, DuBois RN: Cyclooxygenase-2 expression in human colon cancer cells increases metastatic potential. Proc Natl Acad Sci USA 94:3336-3340, 1997 12. Liu XH, Kirschenbaum A, Yao S, et al: Upregulation of vascular endothelial growth factor by cobalt chloride-simulated hypoxia is mediated by persistent induction of cyclooxygenase-2 in a metastatic human prostate cancer cell line. Clin Exp Metastasis 17:687-694, 1999 13. Duffy CP, Elliott CJ, O’Connor RA, et al: Enhancement of chemotherapeutic drug toxicity to human tumour cells in vitro by a subset of nonsteroidal anti-inflammatory drugs (NSAIDs). Eur J Cancer 34:1250-1259, 1998 14. Altorki NK, Keresztes RS, Port JL, et al: Celecoxib, a selective cyclo-oxygenase-2 inhibitor, enhances the response to preoperative paclitaxel and carboplatin in early-stage non-small-cell lung cancer. J Clin Oncol 21:2645-2650, 2003 15. Takahashi T, Ueda S, Ebisawa M, et al: Cyclooxygenase-2 (COX-2) inhibition with meloxicam in combination with paclitaxel and carboplatin in advanced non-small-cell lung cancer (NSCLC). J Clin Oncol 23:683, 2004 16. Johnson DH, Csiki I, Gonzalez A, et al: Cyclooxygenase-2 (COX-2) inhibition in non-smallcell lung cancer (NSCLC): Preliminary results of a phase II trial. J Clin Oncol 23:181, 2004 17. Nugent FW, Mertens WC, Graziano S, et al: Docetaxel and cyclooxygenase-2 inhibition with celecoxib for advanced non-small cell lung cancer progressing after platinum-based chemotherapy: A multicenter phase II trial. Lung Cancer 48:267-273, 2005 18. Sato T, Nakajima H, Fujio K, et al: Enhancement of prostaglandin E2 production by epidermal growth factor requires the coordinate activation of cytosolic phospholipase A2 and cyclooxygenase 2 in

human squamous carcinoma A431 cells. Prostaglandins 53:355-369, 1997 19. Yucel-Lindberg T, Ahola H, Carlstedt-Duke J, et al: Involvement of tyrosine kinase on cyclooxygenase expression and postaglandin E2 production in human gingival fibroblasts stimulated with interleukin-1beta and epidermal growth factor. Biochem Biophys Res Commun 257:528-532, 1999 20. Chen Z, Zhang X, Li M, et al: Simultaneously targeting epidermal growth factor receptor tyrosine kinase and cyclooxygenase-2, an efficient approach to inhibition of squamous cell carcinoma of the head and neck. Clin Cancer Res 10:5930-5939, 2004 21. Tortora G, Caputo R, Damiano V, et al: Combination of a selective cyclooxygenase-2 inhibitor with epidermal growth factor receptor tyrosine kinase inhibitor ZD1839 and protein kinase A antisense causes cooperative antitumor and antiangiogenic effect. Clin Cancer Res 9:15661572, 2003 22. Torrance CJ, Jackson PE, Montgomery E, et al: Combinatorial chemoprevention of intestinal neoplasia. Nat Med 6:1024-1028, 2000 23. Yanagisawa K, Shyr Y, Xu BJ, et al: Proteomic patterns of tumour subsets in non-small-cell lung cancer. Lancet 362:433-439, 2003 24. Shyr Y, Kim K: Weighted flexible compound covariate method for classifying microarray data, in Berrar DP, Dubitzky W, Granzow M (eds): A Practical Approach to Microarray Data Analysis. Boston, MA, Kluwer Academic Publishers, 2002, pp 186-200 25. Thatcher N, Chang A, Parikh P, et al: Gefitinib plus best supportive care in previously treated patients with refractory advanced non-small-cell lung cancer: Results from a randomised, placebocontrolled, multicentre study (Iressa Survival Evaluation in Lung Cancer). Lancet 366:1527-1537, 2005 26. Bombardier C, Laine L, Reicin A, et al: Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis: VIGOR Study Group. N Engl J Med 343:1520-1528, 2000 JOURNAL OF CLINICAL ONCOLOGY



27. Lisse JR, Perlman M, Johansson G, et al: Gastrointestinal tolerability and effectiveness of rofecoxib versus naproxen in the treatment of osteoarthritis: A randomized, controlled trial: ADVANTAGE Study Group. Ann Intern Med 139:539-546, 2003 28. Bresalier RS, Sandler RS, Quan H, et al: Cardiovascular events associated with rofecoxib in a colorectal adenoma chemoprevention trial. N Engl J Med 352:1092-1102, 2005 29. Topol EJ: Failing the Public Health - Rofecoxib, Merck, and the FDA. N Engl J Med 351:1707-1709, 2004 30. Lenzer J: FDA advisers warn: COX 2 inhibitors increase risk of heart attack and stroke. BMJ 330: 440, 2005 31. Shepherd FA, Pereira JR, Ciuleanu T, et al: Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 353:123-132, 2005 32. Simon GR, Ruckdeschel JC, Williams C, et al: Gefitinib (ZD1839) in previously treated advanced non-small-cell lung cancer: Experience from a single institution. Cancer Control 10:388-395, 2003 33. Miller VA, Kris MG, Shah N, et al: Bronchioloalveolar pathologic subtype and smoking history

predict sensitivity to gefitinib in advanced non-smallcell lung cancer. J Clin Oncol 22:1103-1109, 2004 34. Paez JG, Ja¨nne PA, Lee JC, et al: EGFR mutations in lung cancer: Correlation with clinical response to gefitinib therapy. Science 304:14971500, 2004 35. Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 350:2129-2139, 2004 36. Hirsch FR, Varella-Garcia M, Bunn PA, et al: Molecular analysis of EGFR gene copy number, EGFR expression and Akt activation status in patients with advanced non-small-cell lung cancer (aNSCLC) treated with gefitinib or placebo (ISEL trial). Presented at the AACR-NCI-EORTC Symposium, Miami Beach, FL, USA, November 14-18, 2005 37. Gadgeel SM, Shehadeh NJ, Ruckdeschel JC, et al: Gefitinib and celecoxib in patients with platinum refractory non-small cell lung cancer (NSCLC). J Clin Oncol 23:636, 2004 (suppl)

38. Agarwala AK, Einhorn L, Fisher W, et al: Gefitinib plus celecoxib in chemotherapy-naı¨ve patients with stage IIIB/IV non-small cell lung cancer (NSCLC): A phase II study from the Housier oncology group. J Clin Oncol 24:380, 2006 (suppl; abstr 7066) 39. Reckamp KL, Krysan K, Morrow JD, et al: A phase I trial to determine the optimal biological dose of celecoxib when combined with erlotinib in advanced non-small cell lung cancer. Clin Cancer Res 12:3381-3388, 2006 40. Solomon B, Gregorc V, Taguchi F, et al: Prediction of clinical outcome in non-small cell lung cancer (NSCLC) patients treated with gefitinib using Matrix-Assisted Laser Desorption/Ionisation-Time of Flight Mass Spectroscopy (MALDI-TOF MS) of serum. J Clin Oncol 24:365, 2006 (suppl; abstr 7004) 41. Taguchi F, Solomon B, Gregorc V, et al: Mass spectrometry to classify non–small-cell lung cancer patients for clinical outcome after treatment with epidermal growth factor receptor tyrosine kinase inhibitors: A multicohort cross-institutional study. J Natl Cancer Inst 99:838-846, 2007

■ ■ ■

Acknowledgment We thank Maxine Holland, Complete Medical Communications, for medical writing support funded by AstraZeneca.

3273

www.jco.org
