- Email: [email protected]
O-240 Clinical experience with gefitinib (‘Iressa’, ZD1839): an overview of safety and tolerability
s70
Oral Sessions/Novel Therapy: Insights into EGFR Targeted Therapy
feet. We did not obsarva significant variations of treatment effect according to the modalities of CT or RT. Conclusion: Concomitant cisplatin or carboplatin-based chemoradiotherapy may improve survival of patients with locally advanced NSCLC. However, the available data are insufficient to accurately define the size of such a potential treatment benefit and the optimum schedule of chemotherapy.
the time of diagnosis. Since 1973, the five year survival of these patients listed in the SEER database has almost doubled from 10% to 18% (p=O.OOOl for trend). Conclusions: Analysis of these 24 studies and the SEER data shows that there has been a significant improvement in the median and 5-year survivals of patients with stage Ill NSCLC over the past 25 years.
0 239
0 237
El
Concomitant chemoradiotherapy cancer: a systematic review
El
in non-small cell lung
Nick P. Rowell’, Noelle F’. O’Rourkez. ’ Kent Oncology Centre, Maidstone, “Beatson Oncology Centre, Glasgow, UK
UK;
Background: The NSCLCCG 1995 meta-analysis of adjuvant chemotherapy in NSCLC demonstrated a 13% reduction in the risk of death in patients receiving chemotherapy in addition to radical radiotherapy. There have been a number of randomised trials of concomitant chemoradiotherapy (CCRT), a treatment combination specifically excluded from the previous meta-analysis, most of which have shown negative results. Methods: We have conducted a literature-based meta-analysis of randomised trials of CCRT versus radiotherapy alone in stage I-III NSCLC. Trials were eligible for inclusion if the radiotherapy and any chemotherapy given prior to or following radiotherapy were the same in both arms. Overall and progression-free survival at 2 years and morbidity were the chosen end-points. The protocol for the review was published in the Cochrane Library. Randomised trials were identified by electronic searching of the Cochrane Clinical Trials Register, Medline and Embase with identification of further studies from references cited in the initial list. Results: Ten randomised studies (including 1635 patients) met the inclusion criteria. In a meta-analysis, there was a significant increase in Z-year overall survival from 21.3% to 26.4% in those receiving CCRT (relative risk 0.93; 95% confidence interval 0.88-0.98). Similarly, an improvement in 2-year local progression-free survival (relative risk 0.86; 95% confidence interval 0.77-0.97) and progression-free survival at any site (relative risk 0.89; 95% confidence interval 0.80-0.98) was also seen in those receiving CCRT. Subgroup analysis showed that improvements in survival predominated in regimes which delivered chemotherapy weekly or daily during radiotherapy. Treatment-related deaths, acute pneumonitis and late lung fibrosis appeared no more frequently with CCRT but the incidence of acute oesophagitis damage was significantly increased. Conclusions: CCRT improves 2-year overall and progression-free survival compared to radical radiotherapy at the expense of an increase in morbidity with best results being seen in the subgroup receiving daily chemotherapy.
I
0 238
The survival of patients treated for stage Ill non-small cell lung cancer in North America has increased during the past 25 years
Greqorv M. Videtic’, Bruce E. Johnson”, Boris Freidlin3, Scott Saxman4, Steven Schilds, Andrew Turrisi Ills, Everett E. Vokes7, Mitchell Machtaya. ’ Dana Farber Cancer Institute/Brigham & Women’s Hospital, Boston, USA; 2 Dana Farber Cancer Institute, Boston, USA; 3Nationa/ lnsitutes of Health, Bethesda, USA; 4 National Cancer Insitute, Bethesda, USA; 5 Mayo C/ink, Scottsdale, USA; 6 Medical University of South Carolina, Charleston, USA; 7 Univ. of Chicago, Chicago, USA; a Univ. of Pennsylvania, Philadelphia, USA
Background: The median and 5-year survivals of patients with non-small cell lung cancer (NSCLC) listed in the Surveillance, Epidemiology, and End Results (SEER) database has increased over past 25 years: median survival from 9.1 months (1973-74) to 10.1 months (1997-98); 5year survival from 10.5% (1973-1974) to 12.9% (199596). In order to determine the trends in survival for patients with stage Ill NSCLC, we examined data from randomized therapeutic clinical trials performed by co-operative groups in North America initiated between 1970-1994 (and completed by 1998). Methods: Thirty phase Ill trials involving 6262 patients met these criteria. Six trials were excluded from the analyses either due to lack of published information (2) prophylactic cranial irradiation was the only study variable (1) or because they were phase II randomized studies (3). Results: The median of median survivals in the control arms in 13 phase III studies initiated between 1970 and 1985 was 9.1 months and increased to 11 .O months between 1986 and 1998 (p
Gemcitabine in the treatment of Non-Small Cell Lung Cancer (NSCLC): A meta-analysis of survival and progression free survival data
T. Le Chevalier’, A. Browns, R. Natale3, G. Scagliott?, J. Vansteenkiste’, J.L. Van Meerbeeck6, R. Rosel17, R.M. Rudds, S. Dansong, N. Thatcher’, J. Schillerio. On behalf of the Gemzar NSCLC Meta-analysis Group. 1 fnstitute Gustave Roussy, France; *M-TAG, London, UK; 3 Cedars-Sinai Medical Center, Los Angeles, USA; 4 Universitti di Torino, Italy; 5 University Hospital Leuven, Belgium; 6 University Hospital DJkzigt, Rotterdam, Netherlands; 7 lnstitut Catala d’oncologia, Barcelona, Spain; 8 St. Bartholomew’s Hospital, London, UK; g Christie Hospital, Manchester; UK; g Christie Hospital Manchester, UK; I0 University of Wisconsin, USA
Introduction:
Significant advances in the treatment of advanced NSCLC prompted a formal evaluation of the efficacy of gemcitabine-platinum combinations, in comparison with standard and emergent (novel) treatments. A literature review defined the primary focus to any randomised trial of gemcitabine plus cisplatin or carboplatin versus a platinum based regimena. Accordingly a meta-analysis of overall survival (OS) and time to progression (TTP) was performed. Methods: A comprehensive search of published and unpublished sources was performed to identify all trials to December 2002. The hazard ratio (HR) was the summary statistic of choice, accounting for censoring and time-toevent. Where not reported or supplied by the investigator survival probabilities were estimated from Kaplan-Meier curves. The pooled HR was produced using a fixed effects meta-analysis. Statistical heterogeneity was addressed with a random effects model where appropriate. Estimation of absolute treatment benefit at one year was also performed. Results: In total, 13 of the 15 potentially eligible trials were included (one trial was excluded due to flawed randomisation and one trial for data unavailability), resulting in a pool of 4556 patients. A total of 17 comparators were analysed: 12 against platinum-based doublets (11 novel agent-based doublets, VC (6) PC (2) PCb (2), DC (1) and one established agent doublet, EC), plus 5 against singlet/triplet agent regimens (MVC/ MIC (4) and C (1)). For OS a significant reduction in mortality in favour of the gemcitabine based arms was observed, HR 0.90 (0.84 to 0.96, p < O.OOl), using the fixed effects model, with an absolute survival improvement of 3.9% at 1 year. There was a significant reduction in TTP in favour of the gemcitabine regimens, HR 0.87 (0.82 to 0.93, p < O.OOl), with the fixed effects model, with an absolute improvement of progression-free survival of 4.2% at 1 year. Conclusion: Overall the results demonstrate a slight but significant improvement in efficacy of gemcitabine plus a platinum agent when compared with platinum based comparators in survival and time to disease progression. aC, cisplatin, Cb, carboplatin, D, docetaxel, E, etoposide, G, gemcitabine, I, ifosfamide, M, mitomycin, P, paclitaxel, V, vinorelbine.
THURSDAY, 14 AUGUST 2003
Novel Therapy: Insights Targeted Therapy I0
240
into EGFR
Clinical experience with gefitinib (‘lressa’, ZD1839): an overview of safety and tolerability
Beverley Forsvthe, Karen Faulkner. AstraZeneca,
Macclesfield,
UK
The orally active EGFR-TKI (epidermal growth factor receptor tyrosine kinase inhibitor), gefitinib (‘lressa’, ZD1839) has been extensively investigated in non-small-cell lung cancer (NSCLC). In total, over 56,000 patients worldwide have received gefitinib, either within clinical trials, as part of an Expanded Access/-compassionate use programme, or post-approval in Japan. In Phase II trials of patients with advanced, previously treated NSCLC, gefitinib has of patients, symptom benefit demonstrated objective responses in -IO-20% in -40% of patients and -30% one-year survival rates. The most frequently reported adverse drug reactions (ADRs) occurring in >20% of patients with gefitinib in Phase I and II trials were gastrointestinal disturbances (primarily mild diarrhoea) and skin reactions (rash, acne, dry skin and pruritus). These
Oral Sessions/Novel Therapy: Insights into EGFR Targeted Therapy events typically occurred within the first month of treatment, were mild to moderate in intensity (CTC grade 1 or 2) reversible and non-cumulative, and rarely led to withdrawal of treatment. In Phase II studies 250 mg/day was better tolerated than 500 mg/day and is therefore the recommended dose; there ware significantly fewer grade 3/4 ADRs at the 250 mg/day dose compared with 500 mg/day. Patients with diarrhoea or skin reactions may be managed with appropriate symptomatic treatment, followed, if necessary, by a brief therapy interruption (514 days). Asymptomatic increases in liver transaminases (mainly CTC grade 1 or 2) have been reported; periodic liver function testing should therefore be considered. Extensive monitoring revealed no evidence of drug-related ophthalmologic toxicity, while ECG and cardiovascular adverse event data did not show any evidence of conduction abnormalities. Outside of clinical trials, interstitial lung disease-type events have been reported; however, the overall incidence of these events is
pzr
Clinical, Pathologic, and Molecular Characteristics NSCLC Patients Sensitive to Gefitinib
of
Vincent A. Miller, Neelam Shah, Mark G. Kris, Jyoti Patel, Ennapadam Venkatraman, Leah BenPorat, Natalie Memoli, Maureen Zakowski, Barbara Pizzo, Leslie Tyson, Memorial Sloan-Kettering Cancer Center, New York Ci@ USA
ploratory objective of this analysis was to estimate the correlation of epidermal growth factor receptor (EGFR) membrane staining intensity (Ml) with the probability of OR. Tumor biopsies taken between diagnosis and start of gefitinib therapy were assessed by immunohistochemistry for EGFR Ml using the DakoCytomation EGFR pharmDx assayTM. A medical pathologist, blinded to outcome, assessed the percentage of tumor cells with 4 levels of intensity: no staining, weak, moderate and strong (0, l+, 2+ and 3+). Patients with tumor sample evaluable for EGFR and OR were analyzed in the intent-to-treat population. The null hypothesis was that Ml is not predictive of clinical outcome. 157 patients were evaluable (IDEAL 1, 87; IDEAL 2, 70) and of these, 26 (17%) had OR. The demographics of the analysis population were representative of each trial: gefitinib 250/500 mg/day, 80/77; M/F, 95/62; adenocarcinoma/squamous/other, 96/35/26; PS O-1/2, 128/29; prior chemotherapy l/2&3, 39/78/40. The best-fit model was determined for each outcome variable (independently by trial and combined) using a full model with 3 terms for staining intensity. For the analysis of OR using data from both trials, a model based on the proportion of cells staining 2+ or 3+ was the best fit. The model slope estimate (-0.0046; 95% Cl: 0.0017/0.0072, p=O.44) was negative and the mean % staining for 2+ or 3+ was similar for patients with (31.3%; 95% Cl: 16.4%/46.3%) and without OR (37.5%; 95% Cl: 31.2%/43.8%). In both trials 15% (5/34) patients had OR with at least 90% EGFR Ml at no detectable level (0). This exploratory analysis did not reveal a consistent association of relevance between EGFR membrane staining and OR to classify advanced NSCLC patients for treatment with gefitinib. Results of similar analyses of EGFR Ml and symptom improvement will also be presented.‘lressa’ is a trademark of the AstraZeneca group of companies. EGFR pharmDx assay is a trademark of DakoCytomation.
0 243
El
Background: Despite the presence
of the epidermal growth factor receptor (EGFR) on virtually all NSCLC, gefitinib induces objective regressions (9%18%) and symptomatic improvement (35%-43%) in a minority of patients. To identify pretreatment characteristics associated with response, we reviewed data on patients treated with gefitinib at our Institution. Methods: Medical records, pathology reports and specimens, and imaging studies, on all 140 patients treated on one of three gefitinib monotherapy trials at our Center, either from one of 2 clinical trials (n=13) or a compassionate use program (n=l27), were analyzed. The clinical, pathologic, and radiographic features of 21 patients with confirmed radiographic partial responses were compared with the 119 others. Univariate analysis was performed for potential prognostic features linked to gefitinib sensitivity including performance status, gender, histology, prior chemotherapy, and smoking history. Multivariate analysis was performed on the significant variables using an exact logistic regression model. We then created an anonymized database linking these clinicopathologic features with molecular characteristics of available tumor specimens on which multiple IHC or sequencing studies could be performed. These studies, chosen both for relevance to EGFR signal transduction and known relationship to carcinogenic effects of tobacco (k-ras, ~53) include EGFR, p-EGFR, p-ERK, p-Akt, PTEN, HER-2, ~27, ~53, and k-ras and are being performed on 18/21 responders and 30 + non-responders. Results: We found adenocarcinoma (vs. other NSCLC, p
I0 242
Gefitinib (‘lressa’, ZD1839) 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
S71
Clinical benefits in patients with advanced non-small-cell lung cancer treated with gefitinib (“lressa”, ZD1839) in the compassionate use program
Pasi A. Janne’ , Sarada Gurubhagavatula’, Joan Lucca’ , Patricia Ostler’, Arthur T. Skarin’, Panos Fidias*, Thomas J. Lynch’, Bruce E. Johnson’. ’ Dana Farber Cancer Institute, Boston, USA; ‘Massachusetts General Hospital, Boston, USA The compassionate use program for gefitinib (“lressa”, ZD1839) in advanced non-small-cell lung cancer (NSCLC) was designed for patients who were refractory to standard chemotherapy, or could not receive other systemic anticancer therapy. Between 12/00 and 4/02 we enrolled 200 patients (M/F, 95/105) with advanced NSCLC (stage IIIB/IV, 5/195) at the Dana Farber Partners Cancer Care affiliated institutions. Histology: adenocarcinoma, 90; unspecified NSCLC, 54; bronchioloalveolar cell carcinoma (BAC), 20; squamous cell-carcinoma, 17; large-cell carcinoma, 19. The median number of prior chemotherapy regimens was 2 (range O-6), and median performance status (PS) was 1 (range O-3). Median follow-up was 13 months (range 2-23~). Twenty-eight patients were never treated (5 withdrew/23 died). 172 patients received treatment with gefitinib; 11 died within the first month of disease progression. For those patients receiving at least one dose of gefitinib, 7/172 (4.1%; 95% Cl, 1.7-8.2%) patients experienced a partial response (PR); all were women, all had adenocarcinoma, and 4/7 had BAC. 60/172 patients (34.9%) had stable disease (SD). of the 156 patients evaluable for toxicity, grade 3 and 4 toxicity was experienced by 71156 (4.5%) and l/156 (0.6%) patients, respectively. 2056 (1.3%) and 11156 (0.6%) patients discontinued therapy due to grade 3 rash and grade 3 nausea, respectively. Median overall survival for all patients was 4.5 months (95% Cl: 4.1-4.9 months), and 1 -year survival was 29%; both were greater for women than for men: 5.0 months (95% Cl: 3.5-6.5) and 38% versus 4.0 months (95% Cl: 3.34.7) and 18%, respectively; p = 0.027, logrank. 5/7 patients (71.4%) who experienced a PR and 35/60 patients (58.3%) with SD developed skin rash, similar to the overall incidence of rash (73/i 56; 46.8%). Survival was improved in patients who developed any grade skin toxicity versus those without skin toxicity (10.0 months (95%: Cl: 7.4-12.6) versus 4.5 months (95% Cl: 4.1-4.9), respectively; p
I
0 244
Quantitative Gene Expression in Non-Small Cell Lung Cancer from Paraffin-Embedded Tissue Specimens: Predicting Response to Gefitinib, an EGFR Kinase Inhibitor
Renee Bailey’, Mark Kriss, Michael Wolf’, Andrea Kay’, Steven Averbuch’ , Jon Askaa3, Mette Janas3, Karsten Schmidt4, Masahiro Fukuokas. ’Astrai’eneca, Wilmington, USA; 2 Memorial Sloan-Kettering Cancer Center, New York, USA; 3 DakoCytomation, Copenhagen, Denmark; 4 Spadille Aps, Fredensborg, Denmark: 5 Kinki University School of Medicine, Osaka, Japan
Ronald B. Natale. Cedars-Sinai Care, Inc., Los Angeles, USA
In pretreated patients with advanced non-small-cell lung cancer (NSCLC) receiving gefitinib (‘lressa’, ZD1839), objective tumor response (OR; complete + partial response) was 19% (IDEAL 1) and 10% (IDEAL 2). The primary ex-
Purpose: To explore the correlation between quantitative gene expression in tumor tissue, using a high-throughput RT-PCR assay, and response to Gefitinib monotherapy in patients (pts) with non-small cell lung cancer.
Comprehensive
Cancer Center/Salick
Health
Oral Sessions/Novel Therapy: Insights into EGFR Targeted Therapy
feet. We did not obsarva significant variations of treatment effect according to the modalities of CT or RT. Conclusion: Concomitant cisplatin or carboplatin-based chemoradiotherapy may improve survival of patients with locally advanced NSCLC. However, the available data are insufficient to accurately define the size of such a potential treatment benefit and the optimum schedule of chemotherapy.
the time of diagnosis. Since 1973, the five year survival of these patients listed in the SEER database has almost doubled from 10% to 18% (p=O.OOOl for trend). Conclusions: Analysis of these 24 studies and the SEER data shows that there has been a significant improvement in the median and 5-year survivals of patients with stage Ill NSCLC over the past 25 years.
0 239
0 237
El
Concomitant chemoradiotherapy cancer: a systematic review
El
in non-small cell lung
Nick P. Rowell’, Noelle F’. O’Rourkez. ’ Kent Oncology Centre, Maidstone, “Beatson Oncology Centre, Glasgow, UK
UK;
Background: The NSCLCCG 1995 meta-analysis of adjuvant chemotherapy in NSCLC demonstrated a 13% reduction in the risk of death in patients receiving chemotherapy in addition to radical radiotherapy. There have been a number of randomised trials of concomitant chemoradiotherapy (CCRT), a treatment combination specifically excluded from the previous meta-analysis, most of which have shown negative results. Methods: We have conducted a literature-based meta-analysis of randomised trials of CCRT versus radiotherapy alone in stage I-III NSCLC. Trials were eligible for inclusion if the radiotherapy and any chemotherapy given prior to or following radiotherapy were the same in both arms. Overall and progression-free survival at 2 years and morbidity were the chosen end-points. The protocol for the review was published in the Cochrane Library. Randomised trials were identified by electronic searching of the Cochrane Clinical Trials Register, Medline and Embase with identification of further studies from references cited in the initial list. Results: Ten randomised studies (including 1635 patients) met the inclusion criteria. In a meta-analysis, there was a significant increase in Z-year overall survival from 21.3% to 26.4% in those receiving CCRT (relative risk 0.93; 95% confidence interval 0.88-0.98). Similarly, an improvement in 2-year local progression-free survival (relative risk 0.86; 95% confidence interval 0.77-0.97) and progression-free survival at any site (relative risk 0.89; 95% confidence interval 0.80-0.98) was also seen in those receiving CCRT. Subgroup analysis showed that improvements in survival predominated in regimes which delivered chemotherapy weekly or daily during radiotherapy. Treatment-related deaths, acute pneumonitis and late lung fibrosis appeared no more frequently with CCRT but the incidence of acute oesophagitis damage was significantly increased. Conclusions: CCRT improves 2-year overall and progression-free survival compared to radical radiotherapy at the expense of an increase in morbidity with best results being seen in the subgroup receiving daily chemotherapy.
I
0 238
The survival of patients treated for stage Ill non-small cell lung cancer in North America has increased during the past 25 years
Greqorv M. Videtic’, Bruce E. Johnson”, Boris Freidlin3, Scott Saxman4, Steven Schilds, Andrew Turrisi Ills, Everett E. Vokes7, Mitchell Machtaya. ’ Dana Farber Cancer Institute/Brigham & Women’s Hospital, Boston, USA; 2 Dana Farber Cancer Institute, Boston, USA; 3Nationa/ lnsitutes of Health, Bethesda, USA; 4 National Cancer Insitute, Bethesda, USA; 5 Mayo C/ink, Scottsdale, USA; 6 Medical University of South Carolina, Charleston, USA; 7 Univ. of Chicago, Chicago, USA; a Univ. of Pennsylvania, Philadelphia, USA
Background: The median and 5-year survivals of patients with non-small cell lung cancer (NSCLC) listed in the Surveillance, Epidemiology, and End Results (SEER) database has increased over past 25 years: median survival from 9.1 months (1973-74) to 10.1 months (1997-98); 5year survival from 10.5% (1973-1974) to 12.9% (199596). In order to determine the trends in survival for patients with stage Ill NSCLC, we examined data from randomized therapeutic clinical trials performed by co-operative groups in North America initiated between 1970-1994 (and completed by 1998). Methods: Thirty phase Ill trials involving 6262 patients met these criteria. Six trials were excluded from the analyses either due to lack of published information (2) prophylactic cranial irradiation was the only study variable (1) or because they were phase II randomized studies (3). Results: The median of median survivals in the control arms in 13 phase III studies initiated between 1970 and 1985 was 9.1 months and increased to 11 .O months between 1986 and 1998 (p
Gemcitabine in the treatment of Non-Small Cell Lung Cancer (NSCLC): A meta-analysis of survival and progression free survival data
T. Le Chevalier’, A. Browns, R. Natale3, G. Scagliott?, J. Vansteenkiste’, J.L. Van Meerbeeck6, R. Rosel17, R.M. Rudds, S. Dansong, N. Thatcher’, J. Schillerio. On behalf of the Gemzar NSCLC Meta-analysis Group. 1 fnstitute Gustave Roussy, France; *M-TAG, London, UK; 3 Cedars-Sinai Medical Center, Los Angeles, USA; 4 Universitti di Torino, Italy; 5 University Hospital Leuven, Belgium; 6 University Hospital DJkzigt, Rotterdam, Netherlands; 7 lnstitut Catala d’oncologia, Barcelona, Spain; 8 St. Bartholomew’s Hospital, London, UK; g Christie Hospital, Manchester; UK; g Christie Hospital Manchester, UK; I0 University of Wisconsin, USA
Introduction:
Significant advances in the treatment of advanced NSCLC prompted a formal evaluation of the efficacy of gemcitabine-platinum combinations, in comparison with standard and emergent (novel) treatments. A literature review defined the primary focus to any randomised trial of gemcitabine plus cisplatin or carboplatin versus a platinum based regimena. Accordingly a meta-analysis of overall survival (OS) and time to progression (TTP) was performed. Methods: A comprehensive search of published and unpublished sources was performed to identify all trials to December 2002. The hazard ratio (HR) was the summary statistic of choice, accounting for censoring and time-toevent. Where not reported or supplied by the investigator survival probabilities were estimated from Kaplan-Meier curves. The pooled HR was produced using a fixed effects meta-analysis. Statistical heterogeneity was addressed with a random effects model where appropriate. Estimation of absolute treatment benefit at one year was also performed. Results: In total, 13 of the 15 potentially eligible trials were included (one trial was excluded due to flawed randomisation and one trial for data unavailability), resulting in a pool of 4556 patients. A total of 17 comparators were analysed: 12 against platinum-based doublets (11 novel agent-based doublets, VC (6) PC (2) PCb (2), DC (1) and one established agent doublet, EC), plus 5 against singlet/triplet agent regimens (MVC/ MIC (4) and C (1)). For OS a significant reduction in mortality in favour of the gemcitabine based arms was observed, HR 0.90 (0.84 to 0.96, p < O.OOl), using the fixed effects model, with an absolute survival improvement of 3.9% at 1 year. There was a significant reduction in TTP in favour of the gemcitabine regimens, HR 0.87 (0.82 to 0.93, p < O.OOl), with the fixed effects model, with an absolute improvement of progression-free survival of 4.2% at 1 year. Conclusion: Overall the results demonstrate a slight but significant improvement in efficacy of gemcitabine plus a platinum agent when compared with platinum based comparators in survival and time to disease progression. aC, cisplatin, Cb, carboplatin, D, docetaxel, E, etoposide, G, gemcitabine, I, ifosfamide, M, mitomycin, P, paclitaxel, V, vinorelbine.
THURSDAY, 14 AUGUST 2003
Novel Therapy: Insights Targeted Therapy I0
240
into EGFR
Clinical experience with gefitinib (‘lressa’, ZD1839): an overview of safety and tolerability
Beverley Forsvthe, Karen Faulkner. AstraZeneca,
Macclesfield,
UK
The orally active EGFR-TKI (epidermal growth factor receptor tyrosine kinase inhibitor), gefitinib (‘lressa’, ZD1839) has been extensively investigated in non-small-cell lung cancer (NSCLC). In total, over 56,000 patients worldwide have received gefitinib, either within clinical trials, as part of an Expanded Access/-compassionate use programme, or post-approval in Japan. In Phase II trials of patients with advanced, previously treated NSCLC, gefitinib has of patients, symptom benefit demonstrated objective responses in -IO-20% in -40% of patients and -30% one-year survival rates. The most frequently reported adverse drug reactions (ADRs) occurring in >20% of patients with gefitinib in Phase I and II trials were gastrointestinal disturbances (primarily mild diarrhoea) and skin reactions (rash, acne, dry skin and pruritus). These
Oral Sessions/Novel Therapy: Insights into EGFR Targeted Therapy events typically occurred within the first month of treatment, were mild to moderate in intensity (CTC grade 1 or 2) reversible and non-cumulative, and rarely led to withdrawal of treatment. In Phase II studies 250 mg/day was better tolerated than 500 mg/day and is therefore the recommended dose; there ware significantly fewer grade 3/4 ADRs at the 250 mg/day dose compared with 500 mg/day. Patients with diarrhoea or skin reactions may be managed with appropriate symptomatic treatment, followed, if necessary, by a brief therapy interruption (514 days). Asymptomatic increases in liver transaminases (mainly CTC grade 1 or 2) have been reported; periodic liver function testing should therefore be considered. Extensive monitoring revealed no evidence of drug-related ophthalmologic toxicity, while ECG and cardiovascular adverse event data did not show any evidence of conduction abnormalities. Outside of clinical trials, interstitial lung disease-type events have been reported; however, the overall incidence of these events is
pzr
Clinical, Pathologic, and Molecular Characteristics NSCLC Patients Sensitive to Gefitinib
of
Vincent A. Miller, Neelam Shah, Mark G. Kris, Jyoti Patel, Ennapadam Venkatraman, Leah BenPorat, Natalie Memoli, Maureen Zakowski, Barbara Pizzo, Leslie Tyson, Memorial Sloan-Kettering Cancer Center, New York Ci@ USA
ploratory objective of this analysis was to estimate the correlation of epidermal growth factor receptor (EGFR) membrane staining intensity (Ml) with the probability of OR. Tumor biopsies taken between diagnosis and start of gefitinib therapy were assessed by immunohistochemistry for EGFR Ml using the DakoCytomation EGFR pharmDx assayTM. A medical pathologist, blinded to outcome, assessed the percentage of tumor cells with 4 levels of intensity: no staining, weak, moderate and strong (0, l+, 2+ and 3+). Patients with tumor sample evaluable for EGFR and OR were analyzed in the intent-to-treat population. The null hypothesis was that Ml is not predictive of clinical outcome. 157 patients were evaluable (IDEAL 1, 87; IDEAL 2, 70) and of these, 26 (17%) had OR. The demographics of the analysis population were representative of each trial: gefitinib 250/500 mg/day, 80/77; M/F, 95/62; adenocarcinoma/squamous/other, 96/35/26; PS O-1/2, 128/29; prior chemotherapy l/2&3, 39/78/40. The best-fit model was determined for each outcome variable (independently by trial and combined) using a full model with 3 terms for staining intensity. For the analysis of OR using data from both trials, a model based on the proportion of cells staining 2+ or 3+ was the best fit. The model slope estimate (-0.0046; 95% Cl: 0.0017/0.0072, p=O.44) was negative and the mean % staining for 2+ or 3+ was similar for patients with (31.3%; 95% Cl: 16.4%/46.3%) and without OR (37.5%; 95% Cl: 31.2%/43.8%). In both trials 15% (5/34) patients had OR with at least 90% EGFR Ml at no detectable level (0). This exploratory analysis did not reveal a consistent association of relevance between EGFR membrane staining and OR to classify advanced NSCLC patients for treatment with gefitinib. Results of similar analyses of EGFR Ml and symptom improvement will also be presented.‘lressa’ is a trademark of the AstraZeneca group of companies. EGFR pharmDx assay is a trademark of DakoCytomation.
0 243
El
Background: Despite the presence
of the epidermal growth factor receptor (EGFR) on virtually all NSCLC, gefitinib induces objective regressions (9%18%) and symptomatic improvement (35%-43%) in a minority of patients. To identify pretreatment characteristics associated with response, we reviewed data on patients treated with gefitinib at our Institution. Methods: Medical records, pathology reports and specimens, and imaging studies, on all 140 patients treated on one of three gefitinib monotherapy trials at our Center, either from one of 2 clinical trials (n=13) or a compassionate use program (n=l27), were analyzed. The clinical, pathologic, and radiographic features of 21 patients with confirmed radiographic partial responses were compared with the 119 others. Univariate analysis was performed for potential prognostic features linked to gefitinib sensitivity including performance status, gender, histology, prior chemotherapy, and smoking history. Multivariate analysis was performed on the significant variables using an exact logistic regression model. We then created an anonymized database linking these clinicopathologic features with molecular characteristics of available tumor specimens on which multiple IHC or sequencing studies could be performed. These studies, chosen both for relevance to EGFR signal transduction and known relationship to carcinogenic effects of tobacco (k-ras, ~53) include EGFR, p-EGFR, p-ERK, p-Akt, PTEN, HER-2, ~27, ~53, and k-ras and are being performed on 18/21 responders and 30 + non-responders. Results: We found adenocarcinoma (vs. other NSCLC, p
I0 242
Gefitinib (‘lressa’, ZD1839) 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
S71
Clinical benefits in patients with advanced non-small-cell lung cancer treated with gefitinib (“lressa”, ZD1839) in the compassionate use program
Pasi A. Janne’ , Sarada Gurubhagavatula’, Joan Lucca’ , Patricia Ostler’, Arthur T. Skarin’, Panos Fidias*, Thomas J. Lynch’, Bruce E. Johnson’. ’ Dana Farber Cancer Institute, Boston, USA; ‘Massachusetts General Hospital, Boston, USA The compassionate use program for gefitinib (“lressa”, ZD1839) in advanced non-small-cell lung cancer (NSCLC) was designed for patients who were refractory to standard chemotherapy, or could not receive other systemic anticancer therapy. Between 12/00 and 4/02 we enrolled 200 patients (M/F, 95/105) with advanced NSCLC (stage IIIB/IV, 5/195) at the Dana Farber Partners Cancer Care affiliated institutions. Histology: adenocarcinoma, 90; unspecified NSCLC, 54; bronchioloalveolar cell carcinoma (BAC), 20; squamous cell-carcinoma, 17; large-cell carcinoma, 19. The median number of prior chemotherapy regimens was 2 (range O-6), and median performance status (PS) was 1 (range O-3). Median follow-up was 13 months (range 2-23~). Twenty-eight patients were never treated (5 withdrew/23 died). 172 patients received treatment with gefitinib; 11 died within the first month of disease progression. For those patients receiving at least one dose of gefitinib, 7/172 (4.1%; 95% Cl, 1.7-8.2%) patients experienced a partial response (PR); all were women, all had adenocarcinoma, and 4/7 had BAC. 60/172 patients (34.9%) had stable disease (SD). of the 156 patients evaluable for toxicity, grade 3 and 4 toxicity was experienced by 71156 (4.5%) and l/156 (0.6%) patients, respectively. 2056 (1.3%) and 11156 (0.6%) patients discontinued therapy due to grade 3 rash and grade 3 nausea, respectively. Median overall survival for all patients was 4.5 months (95% Cl: 4.1-4.9 months), and 1 -year survival was 29%; both were greater for women than for men: 5.0 months (95% Cl: 3.5-6.5) and 38% versus 4.0 months (95% Cl: 3.34.7) and 18%, respectively; p = 0.027, logrank. 5/7 patients (71.4%) who experienced a PR and 35/60 patients (58.3%) with SD developed skin rash, similar to the overall incidence of rash (73/i 56; 46.8%). Survival was improved in patients who developed any grade skin toxicity versus those without skin toxicity (10.0 months (95%: Cl: 7.4-12.6) versus 4.5 months (95% Cl: 4.1-4.9), respectively; p
I
0 244
Quantitative Gene Expression in Non-Small Cell Lung Cancer from Paraffin-Embedded Tissue Specimens: Predicting Response to Gefitinib, an EGFR Kinase Inhibitor
Renee Bailey’, Mark Kriss, Michael Wolf’, Andrea Kay’, Steven Averbuch’ , Jon Askaa3, Mette Janas3, Karsten Schmidt4, Masahiro Fukuokas. ’Astrai’eneca, Wilmington, USA; 2 Memorial Sloan-Kettering Cancer Center, New York, USA; 3 DakoCytomation, Copenhagen, Denmark; 4 Spadille Aps, Fredensborg, Denmark: 5 Kinki University School of Medicine, Osaka, Japan
Ronald B. Natale. Cedars-Sinai Care, Inc., Los Angeles, USA
In pretreated patients with advanced non-small-cell lung cancer (NSCLC) receiving gefitinib (‘lressa’, ZD1839), objective tumor response (OR; complete + partial response) was 19% (IDEAL 1) and 10% (IDEAL 2). The primary ex-
Purpose: To explore the correlation between quantitative gene expression in tumor tissue, using a high-throughput RT-PCR assay, and response to Gefitinib monotherapy in patients (pts) with non-small cell lung cancer.
Comprehensive
Cancer Center/Salick
Health