Invited Abstracts:

Lung Cancer (2004) 46S1, S1—S30

Invited Abstracts

Update on EGFR inhibitions in NSCLC Paul A. Bunn Jr., M.D. Grohne/Stapp, University of Colorado Cancer Center, Denver, CO, USA The epidermal growth factor (EGFR) has been a potential target for lung cancer therapy since the early 1980s when it was shown that EGF and TGF␣, the EGFR ligands, could stimulate the growth of lung cancer cells expressing EGFR. Multiple studies showed that the EGFR is overexpressed in premalignant bronchial lesions and in the majority of NSCLCs of all histologies. The highest rate of expression was in squamous carcinomas. Expression of EGFR did not predict prognosis, however. Monoclonal antibodies that bind to the EGFR such as C225, were shown to inhibit lung cancer growth in vitro and in athymic mice with human lung cancer xenographs. Small molecule receptor tyrosine kinase inhibitors were subsequently developed and objective responses were noted in the earliest phase I trials. Larger phase II studies confirmed objective responses in 10—20% of advanced NSCLC patients irrespective of the number of prior chemotherapy combinations. Survival of patients in these phase II trials was superior to that reported in historically controlled series using chemotherapy. This survival advantage was recently confirmed in a phase III trial comparing erlotinib to placebo in advanced NSCLC patients who had failed one or more prior chemotherapy regimens. The median survival was improved from — months to — months. The 1-year survival rate from 22% to 31% (HR = 0.72, P < 0.01). Retrospective analysis showed that Asian patients, females, neversmokers and those with adenocarcinoma histology

0169-5002/$ — see front matter doi:10.1016/j.lungcan.2004.07.035

had significantly higher response rates compared to those without these features. The combination of gefitinib or erlotinib plus chemotherapy failed to improve response rates or survival compared to chemotherapy alone. It was not certain whether this failure was related to sequence or whether some patients benefited with synergy while others were harmed by antagonism. A recent retrospective study revealed that response and survival were significantly superior with the combination of erlotinib and chemotherapy in the 10% subset of never smokers. Various biologic features were also studied to determine if they influence sensitivity. Expression of EGFR by immunohistochemistry was not associated with response. However, EGFR gene amplification and high copy number determined by FISH were correlated with response to gefitinib. Several groups have reported that mutations in the kinase domain of the EGFR gene occur in NSCLC patients and that these mutations produce a hyperactive receptor that has increased response to EGF and increased inhibition when exposed to EGFR TKIs. Like the clinical response predictors, these mutations occured more frequently in never-smokers, in adenocarcinomas, in females and in patients of Asian descent. The reasons for these associations are unclear but these are somatic mutations that may be produced by a specific, unidentified carcinogen. At this time it is unclear whether analysis of the EGFR by FISH and mutation analyses will be complementary but it is clear that most responders have both gene amplification and gene mutation. Other studies are in progress to identify those set of genes and set of proteins that may predict sensitivity to EGFR TKIs.

S2 The human and humanized monoclonal antibodies have been studied less in advanced NSCLC patients. It is unclear whether EGFR mutations or amplification predict response though the single agent response rate is also about 10%. Response rates and survival rates after combination therapy with cetuximab and chemotherapy were higher than reported in historical chemotherapy series. One phase II randomized study comparing cetuximab plus chemotherapy to chemotherapy alone showed a higher response rate and a better survival in the combination arm. Randomized phase III trials are in progress to determine if combinations of cetuximab with chemotherapy are superior to chemotherapy alone. The new findings that clinical features and biologic features can select patients for therapy will lead to many new trials using these agents in previously untreated patients. Trials in stages I—III are also in progress. Role of angiogenesis inhibitors in the treatment of lung cancer Joan H. Schiller University of Wisconsin Inhibition of neovascularization as a therapeutic option for cancer patients has become a truism with the publication of two key publications demonstrating the beneficial effects of bevacizumab, a monoclonal antibody to vascular endothelial growth factor. In the first, a randomized, double-blind, phase 2 trial compared placebo with bevacizumab in patients with metastatic renal cell carcinoma, the probability of being progression-free for patients given high-dose antibody, low-dose antibody, and placebo was 64%, 39%, and 20%, respectively, at 4 months and 30%, 14%, and 5% at 8 months [1]. A phase III trial investigated the safety and efficacy of bolus IFL (irinotecan, 5-fluoruracil, leucovorin) with or without bevacizumab in patients with metastatic colorectal cancer. Compared with the FU/LV control arm, treatment with bevacizumab plus IFL resulted in higher response rates (control arm, 35%, combination arm, 45%) and longer median survival (control arm, 15.6 months; combination arm: 20.3 months) [2]. To investigate the efficacy and safety of bevacizumab in lung cancer, a randomized phase II trial was conducted in patients with advanced or recurrent non-small-cell lung cancer. Ninety-nine patients were randomly assigned to bevacizumab 7.5 (n = 32) or 15 mg/kg (n = 35) plus carboplatin and paclitaxel every 3 weeks or carboplatin and paclitaxel alone (n = 32). Compared with the control arm, treatment with carboplatin and paclitaxel

Abstracts plus bevacizumab (15 mg/kg) resulted in a higher response rate (31.5% versus 18.8%), longer median time to progression (7.4 months versus 4.2 months) and a modest increase in survival (17.7 months versus 14.9 months). Of the 19 control patients that crossed over to single-agent bevacizumab, five experienced stable disease, and 1-year survival was 47%. Bleeding was the most prominent adverse event and was associated with squamous cell histology, tumor necrosis and cavitation, and disease location close to major blood vessels [3]. Based upon these results, the Eastern Cooperative Oncology Group has recently completed a randomized Phase III study of bevacizumab (15 mg/kg) plus carboplatin (area under the curve = 6) and paclitaxel (200 mg/m2 ) every 3 weeks or carboplatin and paclitaxel alone. Patients with squamous cell histology were excluded, and cross-over to the experimental arm was not permitted. Over eight hundred patients were entered looking for a 25% improvement in survival. Other intriguing studies involving this agent and lung cancer include a Phase I/II trial of bevacizumab in combination with erlotinib, a EGFr tyrosine kinase inhibitor, for patients with non-small cell lung cancer. A multicenter, Phase I/II study examined the combination in patients with either Stage IIIB/IV or recurrent NSCLC In Phase I, erlotinib 150 mg per day orally plus bevacizumab 15 mg/kg i.v. every 21 days was established as the Phase II dose, although no true dose-limiting toxicities were observed. Forty patients were enrolled and treated, including 34 patients at the Phase II dose level. No pharmacokinetic interaction was observed between these two agents. Seven patients (17.5%) had partial responses, two (5%) had minor responses, and 14 (35%) had stable disease. The median overall survival and time to progression for patients on the Phase II dose level (n = 31) was 9.3 and 4.6 months, respectively [4]. Other anti-angiogenic drugs are also under active investigation and may have activity in lung cancer. Thalidomide is an anti-angiogenic agent which inhibits the expression of VEGF and basic fibroblast growth factor (bVEGF) in preclinical models, and is being studied in an ECOG Phase III study of combined modality therapy for locally advanced NSCLC, with or without thalidomide [5]. SU11248 is tyrosine kinase inhibitor which targets the VEGF receptor and platelet derived growth factor (PDGF) receptor, as well as KIT and flt3, and has been shown to have activity against a variety of human solid tumors [6]. ZD6474 is a potent, oral, low molecular weight TKI of [7] which also has activity versus the tyrosine kinase activity of VEGFR3 and EGFr. Preclinical models demonstrate potent inhibition of

Abstracts VEGF stimulated endothelial cell proliferation in vitro, tumor-induced neovascularization, and significant antitumor activity in a variety of tumor models [7]. BAY 43-9006, originally developed as a cRaf targeted agent, has recently been demonstrated to inhibit other targets, such as VEGFr-2, PDGFr, Flt-3 and c-Kit. Challenges in the future for this group of drugs will include identification of optimal study design, as we attempt to identify surrogate endpoints for Phases I and II studies.

References [1] Yang JC, Haworth L, Sherry RM, Hwu P, Schwartzentruber DJ, Topalian SL, et al. A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody, for metastatic renal cancer. N Engl J Med 2003;349(5):427—34. [2] Hurwitz HI, Fehrenbacher L, et al. Bevacizumab (a monoclonal antibody to vascular endothelial growth factor) prolongs survival in first-line colorectal cancer (crc): results of a phase iii trial of bevacizumab in combination with bolus ifl (irinotecan, 5-fluorouracil, leucovorin) as first-line therapy in subjects with metastatic crc. Am Soc Clin Oncol 2003. [3] Johnson DH, Fehrenbacher L, Novotny WF, Herbst RS, Nemunaitis JJ, Jablons DM, et al. Randomized phase II trial comparing bevacizumab plus carboplatin and paclitaxel with carboplatin and paclitaxel alone in previously untreated locally advanced or metastatic non-small-cell lung cancer. J Clin Oncol 2004;22(11):2184—91. [4] Sandler A, Blumenschein GR, Henderson T, Lee J, Truong M, Kim E, et al. Phase I/II trial evaluating the anti-vegf mab bevacizumab in combination with erlotinib, a her1/egfr-tk inhibitor, for patients with recurrent non-small cell lung cancer. Am Soc Clin Oncol 2004. [5] Li X, Liu X, Wang J, Wang Z, Jiang W, Reed E, et al. Thalidomide down-regulates the expression of vegf and bfgf in cisplatin-resistant human lung carcinoma cells. Anticancer Res 2003;23(3B):2481—7. [6] Mendel DB, Laird AD, Xin X, Louie SG, Christensen JG, Li G, et al. In vivo antitumor activity of su11248, a novel tyrosine kinase inhibitor targeting vascular endothelial growth factor and platelet-derived growth factor receptors: Determination of a pharmacokinetic/pharmacodynamic relationship. Clin Cancer Res 2003;9(1):327—37. [7] Wedge SR, Ogilvie DJ, Dukes M, Kendrew J, Chester R, Jackson JA, et al. Zd6474 inhibits vascular endothelial growth factor signaling, angiogenesis, and tumor growth following oral administration. Cancer Res 2002;62(16):4645—55.

S3 lated and genome related) which may allow for better individualization of therapy. The first trial of targeted therapy, utilizing tamoxifen as a pkc modulator was conducted by CALGB almost 20 years ago. More recently, the ongoing trial CALGB 30203 is the first cooperative group trial to evaluate eicosanoid modulators in advanced NSCLC as well as to evaluate the combination of “targeted agents”. Correlative markers including the expression of target enzymes (COX2 and 5LOX) as well as the target product (VEGF) are required. Trials are currently planned to evaluate the role of dose dense therapy utilizing novel supportive care strategies (CALGB 30303) as well as novel antitubulin agents in the first line setting. CALGB 9633 demonstrated the continuing importance of platinum in therapy of advanced NSCLC. Importantly, as there was a prospectively defined subpopulation analysis for the elderly and PS = 2 patients, the trial demonstrated that these subgroups could both be safely treated as well as demonstrate clear benefit from therapy. CALGB is continuing this emphasis on special populations with a number of studies in development for PS = 2 and/or the elderly. An upcoming trial will evaluate combinations of bortezomib or cetuximab with docetaxel in this setting. QOL studies will be an integral part of these efforts. With the growing population of patients eligible for second and subsequent lines of therapy, CALGB has initiated a number of studies in these areas. CALGB 30401 will evaluate the role of bevacuzimab in conjunction with EGFR inhibitors. Other studies involving novel agents are in development. Bronchoalveolar carcinoma has increasingly been identified as a distinct subset of lung cancer characterized by a distinct epidemiology, biology and response to therapy. CALGB will join SWOG in its study of a dual erb pathway inhibitor as well as develop subsequent trials of novel targeted agents. Integral to all of these studies will be quality of life studies, as well as investigations into the biology of disease including determinants of response and toxicity.

CALGB trials in advanced non-small cell lung cancer M.J. Edelman University of Maryland Greenebaum Cancer Center, Baltimore, MD, USA

EGFR tyrosine kinase inhibitors in advanced nonsmall cell lung cancer: who, what, when, and why Mark G. Kris Memorial Sloan-Kettering Cancer Center, New York, NY, USA

Numerous new agents with activity in advanced NSCLC have been introduced in the past 15 years. CALGB has conducted and continues to conduct hypothesis-based trials directed at improving survival and identifying markers (clinical, tumor re-

The epidermal growth factor receptor tyrosine kinase inhibitors (EGFR TKIs) gefitinib and erlotinib induce dramatic regressions that improve cancerrelated symptoms and survival in persons with

S4 advanced NSCLC. However, most patients do not benefit. Reviewing pretreatment characteristics in clinical trial data reveals that persons who never smoked cigarettes have the highest likelihood of benefit, with a 21%—28% improvement in rates of partial response compared to smokers. Other factors that predict response are female gender, adenocarcinoma (particularly bronchioloalveolar carcinoma) histology, and Asian race. The number or type of prior treatments and the presence of EGFR protein measured by immunohistochemistry do not correlate with outcomes. Mutations in exons 19 and 21 of the EGFR gene in tumor specimens are the strongest predictors of response and likely represent the basis for sensitivity to EGFR TKIs. These mutations have been reported in 81% of 32 patients with partial responses to gefitinib or erlotinib and in 0 of 31 individuals who did not achieve a partial response. Of clinical characteristics that are associated with the presence of a mutation in the EGFR gene, again, being a never smoker tops the list. Among patients sensitive to EGFR TKIs who have mutations in exons 19 or 21 of the EGFR gene in their tumor specimens, 15 of 21 (71%) were never smokers. In an unselected series of surgical NSCLC cases, we found EGFR mutations in 6 of 14 (43%) never smokers and 4 of 81 (5%) current or former smokers. The available data suggests that never smokers have the highest incidence of EGFR mutations as well as the highest likelihood of benefit from EGFR TKIs. In absolute numbers, response rates in never smokers generally exceed those observed both with combination chemotherapy first-line and with single agents after relapse. Therefore, for advanced NSCLC patients who have never smoked cigarettes, gefitinib or erlotinib should be given as initial therapy. For current or former smokers, those with two positive prognostic factors (women, bronchioloalveolar carcinoma, asian race) should also receive these agents initially. For smokers with one or none of the positive predictors, EGFR TKIs should be tried only after chemotherapy, which has a much higher likelihood of benefit. Miller et al. (Proc ASCO 2004) reported an impressive 13month absolute improvement in median survival (23 months versus 10 months) among never smokers randomized to receive chemotherapy plus erlotinib compared to chemotherapy alone. Based on this observation, it appears reasonable to treat never smokers and individuals with EGFR mutations with both chemotherapy and an EGFR TKI. Lastly, analogous to the experience in patients with GIST tumors who, after progression following response to the TKI imatinib, experience even more rapid tumor growth when imatinib is discontinued, it is likely that EGFR TKIs should not be discontinued

Abstracts in individuals with NSCLC with progression following an initial partial response. Recommendations for the use of erlotinib and gefitinib will continue to evolve as our understanding of the molecular events that underlie sensitivity and resistance to these agents grows. The management of advanced NSCLC in special populations: elderly and PS 2 Rogerio C. Lilenbaum, M.D., FACP Thoracic Oncology Program, University of Miami School of Medicine, The Mount Sinai Comprehensive Cancer Center, Miami Beach, FL, USA Elderly patients, defined as 70 years of age or older, represent approximately 40% of all patients diagnosed with non-small cell lung cancer in the US. While the benefit of systemic chemotherapy in younger patients is well recognized, there remains a great deal of skepticism with respect to older patients, who are often labeled as “unfit for chemotherapy”. The landmark “ELVIS” trial was the first to document a clear benefit for chemotherapy in older patients [1]. This study showed a significant survival advantage in favor of single-agent vinorelbine compared to supportive care alone. Importantly, quality of life was also superior in patients treated with vinorelbine. A subsequent trial by the same group (“MILES”) compared single agent vinorelbine and single agent gemcitabine to the combination of the two agents [2]. In this study, there was no advantage for the combination. In the US, Dr. Corey Langer published a subset analysis of elderly patients who were entered in the ECOG trial 5592, which compared cisplatin-etoposide to cisplatin-taxol in two different dosages and schedule [3]. The elderly subset, which comprised 15% of the 574 eligible patients, experienced a higher rate of leukopenia and neuropsychiatric complications, but had no difference in overall quality of life. Of significance, the response rate and survival of elderly patients were comparable to the overall study population. Similarly, Dr. Karen Kelly analyzed the elderly patients entered in two SWOG trials in advanced NSCLC [4]. About 19% of the 605 patients were 70 or older. There was a higher rate of grade 3—5 toxicities, but no significant difference in survival compared to the younger population. A more recent study by the CALGB compared carboplatin-paclitaxel to paclitaxel alone [5]. Of the 561 eligible patients, 155 or 27% were elderly, the highest percentage entered into a cooperative group trial in the US. There was no significant difference in outcome between elderly and younger

Abstracts patients. In addition, toxicity was mild and there was no detriment in quality of life in elderly patients treated with combination chemotherapy. The available data suggests that patients should be evaluated for chemotherapy based on performance status and co-morbidities rather than age. In other words, with the exception of the Italian experience, the data from the US studies supports the use of a carboplatin-based doublet in elderly patients who are judged to be fit to receive combination chemotherapy. Performance status is the most powerful prognostic factor in advanced NSCLC. The notion that patients with PS 2 neither tolerated combination chemotherapy nor benefited from it prevailed throughout the last two decades) and PS 2 patients have been largely excluded form US cooperative group trials [6]. The management of these patients in clinical practice has been inconsistent, ranging from supportive care to single-agent therapy and more recently combination chemotherapy. A subset analysis of ECOG 1595, a four-arm randomized trial of platinum-based doublets, analyzed outcomes for 68 PS 2 patients enrolled into the study [7]. The accrual of PS 2 patients was subsequently suspended due to a high incidence of grade 4/5 toxicities. However, upon further analysis, it became clear that overall toxicity was not significantly different than that observed in the PS 0—1 population. In fact, two of the five deaths on study were attributed to therapy and the remaining three were clearly disease-related. The median survival for this group of patients was 4.1 months and the 1-year survival rate was 19%. The CALGB study discussed above [5] also included a subset analysis of PS 2 patients. Of the 561 eligible patients, 99 (18%) had a PS of 2. Overall, the median and 1-year survival rates in this subgroup were significantly poorer than the PS 0—1 population: 3.0 months and 14%, respectively, compared to 8.8 months and 38% in the PS 0—1 patients. However, and importantly, PS 2 patients treated with the combination regimen had a significantly superior median and 1-year survival compared to those who received single-agent paclitaxel. Actually, the median survival doubled from 2.4 to 4.7 months and the 1-year survival increased from 10% to 18% for those patients treated with the combination. A subsequent study by the ECOG (1599) randomized 98 eligible patients with PS 2 and no prior chemotherapy to attenuated doses of carboplatinpaclitaxel and cisplatin-gemcitabine [7]. Efficacy parameters were comparable: median survival was approximately 6 months in both arms with 1-year survival rates of 20%—25%. Toxicity was acceptable.

S5 Further studies are required to define the optimal management of patients with PS 2. There is some evidence that chemotherapy may prolong survival and improve disease-related symptoms in this subgroup of patients.

References [1] The Elderly Lung cancer Vinorelbine Italian Study group. Effects of vinorelbine on quality of life and survival of elderly patients with advanced non-small cell lung cancer. J Natl Cancer Inst 1999;85:365—76. [2] Gridelli C, Perronet F, Gallo C, et al. Chemotherpay for elderly patients with non-small cell lung cancer: The Multicenter Italian Lung Cancer in the Elderly Study (MILES) phase III randomized trial. J Natl Cancer Inst 2003;95:362—72. [3] Langer CJ, Manola J, Bernardo P, et al. Cisplatin-based therapy for elderly patients with advanced non-small cell lung cancer: implications of Eastern Cooperative Oncology Group 5592, a randomized trial. J Natl Cancer Inst 2002;94:173— 81. [4] Kelly K, Giarritta S, Hayes S, et al. Should older patients receive combination chemotherapy for advanced non-small cell lung cancer? An analysis of the Southwest Oncology Trials 9509 and 9308. Proc Am Soc Clin Oncol 2001;329a (abstract 1313). [5] Lilenbaum RC, Herndon J, List M, et al. Single-agent versus combination chemotherapy in advanced non-small cell lung cancer: a CALGB randomized trial of efficacy, quality of life and cost-effectiveness. Proc Am Soc Clin Oncol 2002;1a (abstract 2). [6] Sweeney C, Zhu J, Sandler AB, et al. Outcome of performance status 2 patients in E1594, a phase III trial in metastatic non-small cell lung cancer. Cancer 2001;92:2639—47. [7] Tester WJ, Stephenson P, Langer CJ, et al. ECOG 1599: Randomized phase II study of paclitaxel/carboplatin or gemcitabine/cisplatin in performance status 2 patients with advanced non-small cell lung cancer. PASCO 2004;23:627 (abstract 7055).

Bronchioloalveolar carcinoma: a model for investigating the biology of EGFR inhibition David R. Gandara, M.D., Howard West, M.D., Kari Chansky, M.S., Angela M. Davies, M.D., Derick H.M. Lau, M.D., Ph.D., John Crowley, Ph.D., Paul H. Gumerlock, Ph.D., Fred R. Hirsch, M.D., Wilbur Franklin, M.D., for the Southwest Oncology Group Bronchioloalveolar carcinoma (BAC) is a previously uncommon subset of non-small cell lung cancer (NSCLC) with unique epidemiology, pathology, clinical features, radiographic presentation and natural history compared with other NSCLC subtypes. Recent data suggest that the incidence of BAC is increasing, notably in younger nonsmoking women. Despite reports of prolonged survival after repeated surgical resection of multifocal lesions and slow growth kinetics, advanced bilateral or recurrent diffuse BAC remains incurable, with the vast majority of patients dying of respiratory failure or in-

S6 tercurrent pneumonia within 5 years. Limited data suggest that chemotherapy may yield poor results in BAC. However, anecdotal reports of prolonged complete response to tyrosine kinase inhibitors of the epidermal growth factor receptor (EGFR), a member of the HER (erbB) family, have raised considerable interest in studying this NSCLC subset. Here we present clinical data and preliminary results of correlative science studies analyzing HER pathways from two prospective Southwest Oncology Group (SWOG) clinical trials performed in advanced stage BAC: S9714 testing a 96-h continuous infusion of paclitaxel (Taxol), and S0126 evaluating the small molecule EGFR inhibitor gefitinib (ZD1839 or Iressa). In S9714, paclitaxel therapy in patients with chemonaive advanced stage BAC resulted in a 14% response rate, median survival of 12 months, and a 2-year survival of 26%. By comparison, S0126 treated 142 patients stratified for prior chemotherapy with gefitinib at a dose of 500 mg per day. The response rate is 19% in chemonaive patients, with a median survival of 12 months and a 2-year survival of 39%. Survival under the influence of gefitinib was significantly increased in females, patients who developed gefitinib-induced rash, PS 0—1 patients, and never-smokers. While EGFR protein levels did not predict outcome, high baseline levels of the downstream marker pMAPK were highly predictive of shortened survival (P = 0.009). K-RAS mutations, a potential contributing variable for resistance, has been found in 13/57 (23%) samples tested. Analysis for activating mutations of EGFR is ongoing, but preliminary results show mutations in 6 of the first 14 patient specimens tested (43%). These studies provide a biologic rationale for using BAC as a model for investigating the biology of EGFR inhibition. Antivascular therapy for non-small cell lung cancer: targeting VEGF and its receptor Roy S. Herbst, M.D., Ph.D. The University of Texas, M.D. Anderson Cancer Center, Houston, TX, USA There is a critical need for new therapies to treat non-small cell lung cancer (NSCLC) because current chemotherapy regimens are of limited effectiveness. The role of vascular endothelial growth factor in promoting tumor angiogenesis, in maintaining existing vasculature, and in resistance to traditional therapies, together with its negative prognostic significance in NSCLC, make it an appropriate target for therapy. Bevacizumab (Avastin), a monoclonal antibody directed against vascular endothelial growth factor, has shown promise in treating a number of different cancers.

Abstracts In a recent phase II trial in patients with advanced metastatic NSCLC, the addition of bevacizumab to standard carboplatin/paclitaxel chemotherapy significantly increased the time to progression and increased the response rate when compared with chemotherapy alone. This was particularly impressive in the subset of patients with non-squamous histology. Bevacizumab is generally well tolerated and did not appear to increase the incidence or severity of nausea/vomiting, neuropathy and renal toxicity, which are typically associated with carboplatin/paclitaxel chemotherapy. Adverse events in phase I and II studies included hypertension, thrombosis, proteinuria (with occasional nephrotic syndrome), and epistaxis. Serious tumor-related bleeding episodes (hemoptysis/hematemesis) seem to be the main safety concern in patients with NSCLC, with squamous cell histology as a possible risk factor. Present ongoing studies are under way in NSCLC including (a) a phase II neo-adjuvant study in combination with paclitaxel and carboplatin in patients with stage IB-IIA NSCLC; (b) a phase I/II study of bevacizumab in combination with the epidermal growth factor receptor tyrosine kinase inhibitor agent, Tarceva, in patients with previously treated NSCLC; and (c) an Eastern Cooperative Group randomized phase III study of paclitaxel and carboplatin with/without bevacizumab in patients with previously untreated IIIB (malignant pleural effusion) or metastatic NSCLC. These studies will help to establish the role of bevacizumab in NSCLC. In addition, several small molecule, oral tyrosine inhibitors which target the VEGF receptor are in early stage studies for NSCLC including ZD6474 and AG013736. Predictive markers for response to EGFR inhibitors Fred R. Hirsch, M.D., Ph.D. University of Colorado Cancer Center, Aurora/ Denver, CO, USA Epidermal growth factor receptor (EGFR) Inhibitors have a significant objective effect in 12—19% of NSCLC patients, who previously have failed on chemotherapy, and stable disease will be achieved in a substantial larger proportion of the patients— –about 40% [1]. However, while phase II studies in previous treated patients have been very promising, studies with EGFR inhibitors as first line treatment in combination with chemotherapy have so far not demonstrated any clinical benefit for the patients. A major obstacle currently is the lack of means to determine,who will have a clinical effect of EGFR inhibitors and who will not.

Abstracts DNA mutations: Recently, specific mutations in the EGFR gene related to the tyrosin kinase domain of the receptor were detected and associated with sensitivity to gefitinib (Iressa® ) [2,3]. However, the occurrence (incidence) of these specific mutations seems to vary considerably according to geography and ethnicity. In an unselected NSCLC population these mutations were detected in only 1 out of 61 US NSCLC patients (2%), but in 15 of 58 Japanese patients [3]. Among 9 patients with response to gefitinib 8 of the patients had tumor with this mutation, while no mutations were found in 7 non-responsive tumors [2]. The findings of these mutation were significant and lead to a more direct focused research. The true incidence of these mutations in different populations is currently under investigations in larger cohorts. Whether these mutations alone in the future can be used as predictive markers to predict EGFR sensitivity has to be studied. In phaseII studies of US NSCLC patients about 10—12% had objective response to gefitinib [4]. Thus, it looks like other mechanisms seems also to be involved in the sensitivity for this agent. Furthermore, it remains to be published data whether these mutations also are associated to other EGFR inhibitors. From a clinical point of view detection of the specific DNA mutations in the EGFR gene is not a ready available clinical test, but it requires specific expertise in DNA sequencing and DNA extraction from mostly paraffin embedded material, which is not very easy. Immunohistochemistry: Much optimism has been related to the use of immunohistochemistry (IHC) for prediction of response to EGFR inhibitors. However, studies based on the IDEAL phase II studies with gefitinib did not show any association between EGFR protein expression and drug sensitivity [5]. However, several issues have to be considered. First of all the histopathologial study was done retrospectively based on submitted and selected material. The material might have been severe selected based on tissue availability. The heterogeneity of EGFR expression in NSCLC has never been studied, and a potential bias in the IHC results due to selection of material might exist in such a retrospective analysis. So far, no peer reviewed publications have documented the lack of association between IHC and sensitivity to EGFR inhibitors in NSCLC. Another factor to be considered is the choice of antibody for the IHC examination. Several EGFR antibodies are on the market most likely targeting different epitopes. Thus, it is still an open question whether one antibody is more predictive than others, and finally another caveat is the lack of uniform criteria for defining a “positive” protein EGFR expression. Whether or not EGFR protein expression predict

S7 sensitivity or not have to be solved in future unselected tumor material by using the same antibody and uniform criterias for EGFR scoring. Recently some downstream proteins detected by IHC were found predictive for gefitinib sensitivity. In a study by Cappuzzo et al. [6] expression of phosphorylated AKT aws clearly associated with gefitinib sensitivity indication that pAKT might be an important predictive marker in future studies. Fluorescence in situ hybridization (FISH): While FISH examination for HER2 amplification seems to be a very good predictive marker for Herceptin treatment in breast cancer, it remains to be seen whether this technology can be used for prediction of EGFR inhibitor sensitivity in NSCLC. From previous studies in NSCLC by using FISH for studying EGFR gene abnormalities it was showed that resected patient with increased copy number of the EGFR gene had a tendency to a shorter survival (like HER2 expression in breast cancer) [7]. Results from on-going studies in gefitinib treated NSCLC patients will show whether FISH markers can be used as predictive markers for gefitinib sensitivity. The advantage of the FISH technology is that this technique can be applied on paraffin embedded material and it is currently a well-established technique in many routine laboratories. Gene array studies: So far, no data from highthroughput gene array studies, i.e. Affymetrix have been published, which can be used for prediction of EGFR inhibitor sensitivity. However, several studies are on-going trying to define a marker panel. The role of (RT) PCR needs also to be studied in future studies. Conclusions: Promising results have been obtained with EGFR inhibitors in NSCLC patients. So far, no validated markers have been demonstrated to predict drug sensitivity in larger clinical trials. Detection of specific mutations in the EGFR gene (codon 19—23) related to the tyrosin kinase domain of the receptor associated to gefitinib sensitivity has been a scientific “breakthrough”. However, there are reasons to believe that also other mechanisms and markers may be involved, which needs to be studied in future studies. While the specific EGFR mutations seem to be associated to tumor response, it is still an open question what the relationship between the DNA mutations and “stable disease” is. Furthermore, it is also an open question whether these EGFR mutations also are associated with tumor response to other EWGFR inhibitors. Taking the complexity of the mechanisms of actions into account (which is not even known in details), it is most likely that a panel of different markers—–and even based on different diagnostic platforms—–might be the best predictors of sen-

S8 sitivity to the EGFR inhibitors. Hopefully, ongoing studies might in the near future give a more valid paradigm for selecting NSCLC patients to EGFR inhibition.

References [1] Fukuoka M, Yano S, Giaccone G, et al. 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—46. [2] 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 2004;350. [3] Paez JG, Janne PA, Lee JC, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib. Science. [4] Kris MG, Natale RB, Herbst RS, et al. Efficacy of gefitinib, an inhibitor of the epidermal growth factor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: a randomized trial. JAMA 2003;290(16):2149—58. [5] Bailey LR, Kris M, Wolf M, et al. EGFR membrane staining is not clinical relevant for predicting response in patients receiving gefitinib (Iressa, ZD 1839) monotherapy for pretreated advanced non-small cell lung cancer: IDEAL 1 and 2. Proc Am Assoc Cancer Res 2003;44:1362. [6] Cappuzzo F, Magrini E, Bartolini S, et al. Improved efficacy of gefitinib therapy in phosphor-AKT positive patients with advanced non-small cell lung cancer. Proc Am Soc Clin Oncol 2004;23:196 (abstract 3004). [7] Hirsch FR, Varella-Garcia M, Bunn PA, et al. Epidermal growth factor receptor in non-small cell lung carcinomas: correlation between gene copy number and protein expression and impact on prognosis. J Clin Oncol 2003;21:3798—807.

Dual tyrosine kinase inhibitors Ramaswamy Govindan, M.D. Washington University, St. Louis, MO, USA Several new orally administered drugs that inhibit more than one tyrosine kinase have now become available for clinical trials. In some instances, these “dual inhibitors” block predominantly the activity of (a) epidermal growth factor receptor (EGFr, ErB-1) and ErB-2 signaling pathway or (b) EGFr and vascular endothelial growth factor receptor (VEGFR-2) or (c) VEGFR-2 and platelet derived growth factor (PDGF). GW 572016 is a small molecule tyrosine kinase inhibitor that interrupts EGFr and Erb2 signal transduction pathways. When administered at a dose level of 900—1600 mg a day, it is well tolerated with the most frequent side effects being grade 1—2 rash, anorexia, fatigue, nausea and vomiting. Preliminary evidence indicates anti-tumor activity in heavily pretreated patients with various malignancies including advanced non-small cell lung cancer. ZD6474 inhibits tyrosine kinase activity of VEGFR-2 and EGFr. When administered at a dose of 100—300 mg per day, it is extremely well tolerated. The most common side

Abstracts effects include rash, diarrhea, asymptomatic QTc prolongation, mild proteinuria, and hypertension. In a phase I study conducted in Japan, tumor regression was noticed in patients with non-small cell lung cancer. Phase II studies are ongoing to evaluate the activity of ZD6474 in patients with metastatic nonsmall cell lung cancer. SU11248 is a novel oral drug that targets VEGF-R, PDGF-R, FLT3 and KIT. At the recommended phase II dose of 50 mg a day, the most common side effects observed are grade 2 skin/hair discoloration, mucositis and occasionally edema and thrombocytopenia. Tumor necrosis with intratumoral bleeding has been observed with the use of this compound. SU11248 has produced clinical benefit in patients with gastrointestinal stromal tumor (GIST) refractory to imatinib. Studies are ongoing to evaluate the activity of this agent in neuroendocrine malignancies including those arising in the lung. Role of c-Met receptor tyrosine kinase in lung cancer Patrick C. Ma, Jagadeeswaran Ramasamy, Tanguy Seiwert, Maria S. Tretiakova, Aliya N. Husain, Neelu Puri, Ravi Salgia Pritzker School of Medicine, and University of Chicago Cancer Research Center, University of Chicago, Chicago, IL, USA Receptor tyrosine kinases (RTKs) have been shown to be important in a variety of malignancies, such as HER2/NEU in breast cancer, c-Kit in gastrointestinal stromal tumors and epidermal growth factor receptor in non-small cell lung cancer (NSCLC). In order to impact on future therapies of solid tumors, further studies on RTKs are urgently needed. One such RTK, c-Met, has been shown to be overexpressed in a number of solid tumors, especially lung cancer. c-Met signaling plays a role in growth, motility, invasion, metastasis, angiogenesis, wound healing, and tissue regeneration. The expression of c-Met and its ligand hepatocyte growth factor (HGF) has been determined in breast cancer, ovarian cancer, prostate cancer, colon cancer, melanoma, head and neck cancer, renal cell carcinoma, brain tumors, mesothelioma, lung cancer, and other carcinomas. In particular, we have studied the role of c-Met in lung cancer, especially NSCLC and small cell lung cancer (SCLC). There was variable expression of c-Met in 10 SCLC cell lines, whereas majority of the 9 NSCLC cell lines had c-MET over-expression. c-Met expression was seen also in tumor tissues of lung cancer. Activated-Met expression was demonstrated in all of the tumor types, at varying degrees, using the phospho-specific antibodies against pY1003 (juxtamembrane) and

Abstracts pY1230/1234/1235 (auto-phosphorylation) sites. HGF induced pY1230/1234/1235-Met expression dramatically in A549 cells, as shown by immunofluorescence studies. c-Met expression was successfully inhibited by siRNA technique by >50% in the NSCLC A549 cell lines. We have also identified specific somatic mutations in lung cancer, especially in the juxtamembrane and semaphorin domains. Mutations of the juxtamembrane domain lead to enhanced cell motility and activation of the focal adhesion protein paxillin (especially pY31). Signaling through HGF/c-Met axis activated the focal adhesion signaling molecules (paxillin, FAK, and PYK2), PI3K/Akt/mTOR pathway, as well as the Erk1/2 pathway. We are currently testing small molecule inhibitors, such as SU11274 and PHA665752, against the c-Met pathway. In order to rapidly screen c-Met inhibitors, we have established an in vitro cell line model with BaF3 (pre-B) cells and H446 (SCLC) cells stably transfected with TPR/MET activated oncogene. c-Met receptor pathway appears to be important in lung cancer, and it has an important role as a potential novel therapeutic target. Extensive stage small cell lung cancer Bruce E. Johnson, M.D. Lowe Center for Thoracic Oncology, Dana Farber Cancer Institute and Departments of Medicine, Brigham and Women’s Hospital and Harvard Medical School Patients with extensive stage small cell lung cancer treated with combination chemotherapy including etoposide cisplatin currently achieve a median survival of approximately 10 months. The median, 1- and 2-year survival of patients treated with combination chemotherapy have not substantially changed in the past decade and recent trials have not provide an obvious alternative to the standard etoposide cisplatin regimen. The large randomized phase III trials reported from the North America, Europe, and Japan in the last 5 years have added clarification to some important therapeutic concepts for patients with extensive stage small cell lung cancer. The most encouraging findings have been from Japan published in 2002. The 154 patients with extensive stage small cell lung cancer were randomized to be treated with irinotecan plus cisplatin or etoposide cisplatin. The trial was stopped early because the patients treated with the irinotecan combination had a median survival of 13 months compared to 9 months for those treated with the etoposide combination. There are two confirmatory trials ongoing for patients with extensive stage small cell lung cancer

S9 in the United States. Both trials compare the outcome of patients treated with irinotecan cisplatin to those treated with etoposide cisplatin. One trial sponsored by Pfizer has completed accrual while the other trial being run by the Southwest Oncology Group continues to accrue patients. The results of these trials should be available within the next 1—2 years. If the patients treated with irinotecan cisplatin live longer than those treated with etoposide cisplatin, this will become the new standard regimen for patients with extensive stage small cell lung cancer. The other trials have not shown a consistent improvement in outcome. The addition a third drug to etoposide cisplatin or other three-drug regimens do not improve the survival of patients with extensive stage small cell lung cancer. A trial of 222 patients with extensive stage small cell lung cancer reported by Sundstrom in 2002 showed those treated with etoposide cisplatin had a similar median and overall survival as patients treated with the three-drug regimen, cyclophosphamide, epirubicin, and vincristine. The addition of paclitaxel to the combination etoposide and cisplatin for 293 patients with extensive small cell lung cancer in the Cancer and Acute Leukemia Group B did not show a difference in outcome compared to the 294 patients treated with etoposide cisplatin when reported at ASCO in 2002. The other trials pursued in Europe have mostly tested doseintense strategies. These have not produced consistent results showing improved survival for the patients with extensive stage small cell lung cancer treated with the dose-intense regimens. The addition of maintenance therapy has not been shown to improve outcome as well. Recent trials have reconfirmed that patients with extensive stage small cell lung cancer can be safely treated with single new novel agents before switching to conventional therapy. This design should be used to test new agents to expand the number of agents available for future testing in larger trials for patients with extensive stage small cell lung cancer. Modulating apoptotic signals in small cell lung cancer Charles M. Rudin, Christine L. Hann, Ravi Salgia, Everett Vokes John Hopkins University, Baltimore, MD, USA Small cell lung cancer is an aggressive malignancy associated with rapid growth, high metastatic potential, and poor prognosis, with a median survival of approximately 9 months. Unlike many solid tumors, de novo small cell lung cancer is highly re-

S10 sponsive to genotoxic chemotherapy, and response is associated with apoptotic induction. In contrast, recurrent disease is typically chemoresistant. We have become interested in the modulation of apoptotic signaling pathways as a means of improving outcome for patients with small cell lung cancer, both to increase efficacy of first line chemotherapy, and to reverse chemoresistance in recurrent disease. Potential approaches being considered by our group and others include targeting upstream survival signaling receptors such as IGF-1R, and targeting downstream central apoptotic regulators such as Bcl-2. Small cell lung cancer is not associated with high-level expression of EGFR family members, but rather with a distinct set of receptor tyrosine kinases including IGF-1R and c-kit. Initial results of clinical trials inhibiting c-kit alone with imatinib in small cell lung cancer have been disappointing. Recent data suggests that targeting IGF-1R, with or without inhibition of c-kit, can inhibit proliferation and promote apoptosis in small cell lung cancer, and that combinatorial inhibition may be synergistic. This approach has not yet been tested clinically. We have explored the use of an antisense oligonucleotide directed against Bcl-2, G3139, in patients with small cell lung cancer. Two phase I studies, combining G3139 with paclitaxel in patients with advanced chemorefractory small cell lung cancer, and with carboplatin and etoposide in patients with previously untreated small cell lung cancer, have been completed. A randomized phase II trial evaluating G3139, carboplatin and etoposide is ongoing in the Cancer and Leukemia Group B, and is projected to complete accrual in 2004. In addition to inhibition of Bcl-2 by antisense oligonucleotides, high affinity small molecule inhibitors of Bcl-2 have been developed. A small molecule inhibitor of Bcl2 has been shown to have single agent activity against small cell lung cancer xenografts expressing Bcl-2. Further evaluation of this and related inhibitors in small cell lung cancer xenograft models, either alone or together with standard cytotoxic chemotherapy, is ongoing. Recent definition of key regulators of cell survival expressed in small cell lung cancer offers a unique opportunity for focused novel therapeutic development against this disease. Tumor COX-2 expression: a window to pathways and targets in non-small cell lung cancer Steven M. Dubinett, Harnisha Dalwadi, Kostyantyn Krysan, Mariam Dohadwala, Sherven Sharma, Jenny T. Mao, Mehis Pold, Brian Gardner, Robert Strieter, Robert Figlin, Karen Riedl University of California at Los Angeles Lung Cancer Research Program of the Jonsson Comprehensive

Abstracts Cancer Center, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA Cyclooxygenase (COX-2) is elevated in non-small cell lung cancer (NSCLC). Tumor COX-2 overexpression appears to affect numerous cellular pathways, including tumor invasion, angiogenesis, suppression of immunity, and resistance to apoptosis. However, the precise mechanisms of these cellular changes are presently unclear. We performed Affymetrix microarray (U133A chip) to elucidate COX-2 dependent gene expression in NSCLC. We examined 15,000 genes in A549 (human lung adenocarcinoma cell line) and H157 (squamous cell carcinoma cell line) either over-expressing COX2 or expressing little or no COX-2. 159 genes were differentially expressed (either twofold increase or decease) in A549 cells expressing high COX-2 as compared to A549 cells with no COX2 expression (COX-2 anti-sense). With regards to H157 NSCLC cell line, 2184 genes were differentially expressed (either twofold increase or decease) in H157 cells expressing high COX-2 when compared to COX-2 anti-sense cells. From these two gene subsets, only 48 genes were increased and 11 genes were decreased in both cell lines (A549 and H157) expressing high COX-2. Many of the genes identified in this microarray study are relevant to tumorigenesis and can be classified into various categories such as extracellular matrix, chemotaxis, immune/inflammatory response, and apoptosis. Functional studies evaluating these genes and proteins are defining important downstream mediators of COX-2-dependent regulation of malignant phenotypes in NSCLC. Another related avenue to determine COX-2-dependent contributions to lung cancer progression is through investigation of PGE2-dependent signaling mechanisms that promote tumor growth. Previous reports have linked the COX-2 product PGE2 to the abnormal activation of MAPK/Erk kinase pathway. Here we demonstrate that PGE2 is able to rapidly stimulate Erk phosphorylation in a subset of NSCLC cell lines. This effect is not evident in bronchial epithelial cells. In contrast to previous reports in colon cancer, we found that Erk activation by PGE2 was not inhibited by pre-treatment of the cells with EGF receptor (EGFR) inhibitors, despite the fact that this treatment completely abrogated EGF-induced signaling. Activation of the Erk pathway by PGE2 was also resistant to src kinase inhibitors. PGE2 effects are mediated through four G-protein-coupled receptors—–EP1, EP2, EP3 and EP4. Selective inhibition of EP receptors revealed the possible involvement of Ca2+ -dependent signaling in PGE2mediated activation of Erk. Our data suggest the

Abstracts presence of yet unknown, EGFR-independent crossactivation of MAPK/Erk pathway by PGE2 in lung cancer cells. These findings provide evidence for the possible link between tumor COX-2 overexpression and elevated Erk-mediated lung cancer cell proliferation and migration. This new understanding in turn may lead to significant advances in the use of COX-2 inhibition in combination with other targeted or standard therapies. Based on these findings, a phase I trial of a COX-2 inhibitor (celecoxib) in combination with an EGFR inhibitor (erlotinib) in metastatic NSCLC is underway. Combination of drugs with different molecular targets in cancer treatment Fortunato Ciardiello Division of Medical Oncology, Second University of Naples, Italy In the past 20 years the understanding of the basic molecular alterations which lead to cancer development and progression has allowed an extraordinary evolution in the design of molecular targeted strategies in cancer treatment. The blockade of the functional activity of cell membrane growth factor recepors which play a key role in cancer cell proliferation and in tumor-induced angiogenesis is being developed as a valuable strategy for cancer therapy. For example, selective and potent inhibitors of epidermal growth factor receptor (EGFR) signaling, such as gefitinib, erlotinib and cetuximab have demonstrated antitumor activity in a subset of human epithelial cancers, including non-small cell lung cancer (NSCLC), advanced colorectal cancer and advanced head and neck squamous cell carcinoma (HNSCC) patients. However, recent experimental data suggest that cancer cells may escape from growth inhibition due to blockade of a single signaling pathway, such as the EGFR pathway, by using alternative growth pathways or by constitutive activation of downstream signaling effectors. In fact, it is conceivable that multiple growth-controlling pathways are intrinsically altered or can be activated in cancer cells following treatment with specific inhibitors which are selective for a single growth factor receptor, contributing to the development of acquired resistance to these drugs. Therefore, simultaneous inhibition of multiple molecular targets may be required to obtain an optimal therapeutic effect in human epithelial cancers. A potentially relevant approach is the combined inhibition of a growth factor signaling pathway, such as, for example, the EGFR pathway, in cancer cells and of angiogenesis by targeting directly or indirectly tumor vessels. This strategy seems reasonable for several mechanistic

S11 reasons, including the activation of EGFR by EGF or TGF (which up-regulates the production of proangiogenic growth factors, such as vascular endothelial growth factor (VEGF) by cancer cells. Furthermore, recent preclinical studies have demonstrated that acquired resistance of human cancer xenografts to EGFR inhibitors is associated to the enhanced and dysregulated production of VEGF and, posssibly of other angiogenig growth factors. In this respect, a series of preclinical studies have shown that the combined blockade of the EGFR pathway in cancer cells and of endothelial cell proliferation results in a more efficient and sustained tumor growth control. Therefore, a series of clinical trials are undergoing to evaluate this experimental hypothesis. The most promising strategy has been the combination of a selective EGFR inhibitor, such as the small molecule tyrosine kinase inhibitor erlotinib, and a direct antiangiogenic agent such as the VEGF neutralizing monoclonal antibody bevacizumab. Initial results of this approach from phase I—II clinical studies in NSCLC and HNSCC patients demonstrate that this therapy is feasible and has promising clinical activity. Trimodality therapy for stage IIIA non-small cell lung cancer: updated North American experience Kathy S. Albain, M.D. Thoracic Oncology Program, Stritch School of Medicine, Cardinal Bernardin Cancer Center, Loyola University Chicago, Chicago, USA The International Association for the Study of Lung Cancer (IASLC) recently convened a workshop on the management of non-small cell lung cancer. A committee was charged with providing a consensus report on “Induction Treatment before Surgery”. This report recommended that for advanced stage III disease, chemotherapy plus radiotherapy prescriptions remained the standard treatment approach for patients with good performance status [1]. The role of surgery after induction chemoradiotherapy was still viewed as unsettled for the majority of patients, with randomized trial data awaited. Two exceptions were noted for which trimodality therapy should be strongly considered based on quite convincing phase II data: the superior sulcus presentation and the relatively uncommon T4N0/1 subset. However, for the majority of patients with stage IIIAN2 disease due to high volume mediastinal involvement, bimodality treatment with chemoradiotherapy was recommended as standard care. Since publication of this Consensus Report, the first survival results of the North American Intergroup Trial 0139 were presented and updated [2,3].

S12 This phase III study randomized patients between chemoRT alone versus the same chemoRT followed by surgical resection. The survival results were based on complete accrual, but incomplete surgical details and long-term follow-up data. Progressionfree survival was significantly prolonged in the trimodality arm, but there was no significant overall survival advantage. Crossing hazard ratios were observed due to early mortality in the surgical arm, but subsequent improved progression-free survival. Study coordinator analyses of outstanding data forms were completed in May, 2004, and a final survival analysis is planned soon. Results of this large study will be placed in context of other ongoing North American trials. Based on these data, this presentation will conclude with a consideration of whether the Consensus Statement is outmoded, or instead remains completely accurate regarding a role for trimodality therapy in advanced stage III non-small cell lung cancer.

References [1] Eberhardt et al. Lung Cancer 2003;42:S9—14. [2] Albain et al. Proc Am Soc Clin Oncol 2003. [3] Albain et al. Proceedings of the 10th World Conference on Lung Cancer. Lung Cancer Suppl 2003.

Multimodality approaches to stage IIIA NSCLC—– the European perspective Wilfried Eberhardt1 , Thomas Gauler1 , S¨ onke Korfee1 , Christoph P¨ ottgen2 , Martin Stuschke2 1 Department of Internal Medicine (Cancer Res.), 2 Department of Radiation Oncology, West German Cancer Centre Essen, University of Duisburg-Essen, Germany Background: Stage IIIA represents the subset of NSCLC patients that may profit from interdisciplinary treatment strategies the most. Different permutations of chemotherapy, radiotherapy and surgery have been tested during recent years within different clinical investigations. However, even this sub stage shows a remarkable heterogeneity of disease biology and prognosis that has to be taken into account when analyzing results from individual trials projects. IIIA spans from single, minimal or even microscopic lymph node involvement within the mediastinum to multilevel, extra capsular, lymphangiotic or even bulky mediastinal lymph node disease. For the vast majority of patients, concurrent application of chemotherapy and radiotherapy represents a typical and accepted treatment approach while upfront surgery is usually only performed in the small group of

Abstracts patients with proven minimal mediastinal involvement. Methods: Herein, we will try to review major recently performed, finished or ongoing multicenter European trials projects in stage IIIA disease. Different cooperative groups have tried to generate evidence for the optimal treatment approach to these patients. For this analysis, the main emphasis was laid on reported results from randomized multicenter trials. Results: Evaluating the inclusion of surgery within the patient subset of stage IIIA-disease has been a frequent theme tested in European trials during recent years. (1) The EORTC trial has tested the role of surgery compared to radiation therapy following induction chemotherapy in (non-progressing) patients with mediastinoscopically proven stage IIIA(N2)-disease (Giaccone et al.). This trial has finished patients accrual but only data on perioperative toxicity have been reported yet. A major criticism to this multicenter trial that made a lot of compromises on the choice of chemotherapy combinations included is the (sequential) radiotherapy approach included in the standard arm. Survival data for this study are awaited to be reported next year. (2) The multicenter randomized trial of the German Lung Cancer Cooperative Group (Thomas et al.) has tested the inclusion of radiotherapy either preoperatively or postoperatively within the setting of induction therapy followed by surgery—–if possible—–in patients with locally advanced stages IIIA to IIIB. The results of this large randomized study have been presented at this year’s ASCO. At this time and point of analysis no significant benefit on survival, time to progression or complete resection rate by the inclusion of radiotherapy within the preoperative setting could be substantiated. Criticism to this trial points to the marked heterogeneity of the patient population included, the performance of radiotherapy within both arms of the trial and the probably suboptimal concurrent chemotherapy protocol to radiation in the preop setting. Further follow-up will probably be able to give a more detailed analysis of implications from this trial for individual patients sub sets of IIIA-disease. (3) A multicenter German randomized trial by the West German Lung Cancer Consortium has looked at the small but well defined group of patients with operable (“minimal”) IIIA-disease (Eberhardt et al.). A local-treatment-only approach of surgery followed by postoperative radiotherapy has been randomized against a multimodality protocol of induction chemo-therapy, chemo radiation and definitive surgery. Although terminated early due to slow accrual and the conflicting situation with

Abstracts the IALT results becoming available, more than 110 patients had been randomized till 2001. The results of this randomized study confirmed the efficacy and acceptable toxicity profile of such a complex multimodality protocol also in the multicenter setting. So far, a clear trend for longer survival as well as progression-free survival was observed in the Trimodality arm, although due to the small number of patients no statistical significance could be established. Interestingly, this trial observed for the first time downstaging effects due to the chosen complex induction protocol resulting in a clear trend towards organ sparing surgery in the Trimodality arm. Conclusions and outlook: Taken together the reported data from European trials in stage IIIA NSCLC, surgery still has a role in the treatment of this patient subset. Little data currently exist on the prognosis of the same patients selection treated with modern (and conformal) chemoradiotherapy protocols. Consequently, a multicenter German—French Intergroup trial (GERMAN KREBSHILFE) has started in patients with stage IIIA(N2) and selected IIIB-disease testing the role of surgery versus a definitive chemo radiation boost following a chemotherapy and concurrent chemo-radiation induction protocol. Patients are randomized at the end of induction therapy when being evaluated as “operable” and “resectable” by the multidisciplinary treatment team. Adjuvant chemotherapy improves overall and disease-free survival in non-small cell lung cancer (NSCLC): results of the IALT trial T. Le Chevalier, M. Kozlowski, J.P. Pignon, B. Bergman, T. Orlowski, R. Pirker, C. Le Pechoux, M. Tarayre, A. Dunant, on behalf of the IALT Investigators We recently published the results of a large international adjuvant lung cancer trial (IALT) which was designed to question the role of 3—4 cycles of adjuvant cisplatin-based Ct after complete resection of NSCLC [1]. Each center predetermined the cisplatin dose (total 300—400 mg/m2 ), the combined drug (etoposide or a vinca-alkaloid) and the radiotherapy policy. Tests were two-sided. Analyses were adjusted on center, type of surgery and pathological stage with Cox models. From 1995 to 2000, 1867 patients were randomized in 148 centers from 33 countries. On September 1st, 2002, median follow-up was 56 months and more than 98% of patients had a follow-up up to date. There were 935 patients in the Ct arm and 932 patients in the control arm. Overall survival was significantly different between the two arms: 5-year survival rate was 44.5% in the Ct arm versus 40.4% in the control arm (RR = 0.86 [0.76—0.98], P

S13 < 0.03). Disease-free survival was also significantly different: 39% in the Ct arm versus 34% in the control arm at 5 years (RR = 0.83 [0.74—0.94], P < 0.003). Incidence of local recurrence was significantly different between the two arms: 5-year rate was 24% in the Ct arm versus 29% in the control arm (P < 0.003). Incidence of distant metastasis was significantly different between the two arms: 5-year rate was 41% in the Ct arm versus 44% in the control arm (P < 0.03). Incidence of brain metastases was not significantly different between the two arms: 5-year rate was 16% in the Ct arm and 14% in the control arm (P = 0.64). Incidence of other metastases was significantly different between the two arms: 5-year rate was 29% in the Ct arm versus 35% in the control arm (P < 0.003). Incidence of second cancer was not significantly different between the two arms: 5-year rate was 6% in the Ct arm and 7% in the control arm (P = 0.64). We concluded that adjuvant Cisplatin based Ct in resected NSCLC significantly improve overall and disease-free survival. It decreases both local and distant recurrences and should become part of the standard management of operable NSCLC. Three other randomized prospective studies recently reported a significant benefit of adjuvant modern single agent UFT or doublets in early stage NSCLC and confirmed the role of adjuvant chemotherapy as part of the treatment of resected NSCLC. The LACE program, a pooled analysis of all recent adjuvant trials, and the IALT-bio study should allow to better define the populations more able to benefit from post-operative chemotherapy. Supported mainly by Association pour la Recherche sur le Cancer and Programme Hospitalier de Recherche Clinique). Adjuvant lung project Italy (ALPI): the Italian/EORTC-LCCG randomised trial of adjuvant cisplatin-based chemotherapy in completely resected non-small cell lung cancer (NSCLC) Giorgio Scagliotti on behalf of the ALPI/EORTCLCCG investigators Orbassano, Italy This phase III study tested the hypothesis whether adjuvant cisplatin-based chemotherapy provides better survival than surgery alone in completely resected NSCLC. Patients with stage I, II and IIIa NSCLC were randomised postoperatively to receive either MVP (Mitomycin 8 mg/m2 day 1; Vindesine 3 mg/m2 day 1 and 8; Cisplatin 100 mg/m2 day 1 q 3 weeks for 3 cycles) or no chemotherapy. Delivery of postoperative radiotherapy (total dose 50—54 Gy in 5—6 weeks, beginning at least 4 weeks after the completion of chemotherapy), was left to the policy of participating centres and randomisation was

S14 stratified accordingly. Between January 1994 and February 1998, 1209 pts were randomised, 13 pts were ineligible; of the eligible pts, 602 were allocated to chemotherapy and 594 to control. Main patients’ characteristics were: male = 88%; mean age = 61 years; stage: I = 42%, II = 31%, IIIa = 27%. Radiotherapy was planned in 4%, 60%, 76% of stage I, II, and IIIa pts, respectively. As for chemotherapy compliance, 327 (69%), out of 474 evaluable pts, completed the treatment, 166 of them with modifications of the planned regimen. One hundred (21%) interrupted treatment while 47 pts never started chemotherapy. Overall, 127 (30%) and 79 (18%) MVP pts experienced grade 3 and grade 4 toxicity, respectively. With a median follow up of 63 months, 1076 pts are available for survival analysis, 526 pts have died (260 MVP, 266 control), while 592 have relapsed or died (289 MVP, 303 control). No statistically significant difference in multivariate overall (HR = 0.94; 95% CI 0.79—1.12) or event free (HR = 0.89; 95% CI 0.75—1.04) survival analysis was found. Despite the incidence of grade 3—4 hematological and non-hematological toxicities related to chemotherapy did not differ quantitatively and qualitatively from those commonly reported in advanced NSCLC, the marginal reduction in survival observed in the MVP arm in the first year after randomization may potentially reflect a toxicity effect. This is also indirectly confirmed by the lower percentage of patients in the MVP arm who completed subsequent thoracic radiotherapy (65% versus 81% in the control arm). Another potential issue may be related to the already known pulmonary toxicity of mitomycin C and a potential detrimental interaction with sequential radiotherapy. However, when the causes of the early deaths (occurred within 12 months since randomisation) were analysed most of deaths excess (90 in the chemotherapy arm versus 69 in the control arm) was attributable to cancer progression (11 cases) and other cardiopulmonary events (7 cases) while clearly treatment-related deaths was inversely distributed (3 versus 6). In addition, when mitomycin C, vinblastine and cisplatin regimen was administered concurrently with radiotherapy in unresectable stage III NSCLC no increase in treatmentrelated mortality and morbidity was observed. Moreover, there was not good evidence of differential effect of chemo in different subgroup of patients: in stage I, II and IIIa the hazard ratios for ratios for survival where 0.97 (0.71—1.33), 0.79 (0.60—1.05) and 1.08 (0.83—1.40), respectively (Pvalue for interaction = 0.52). Similar figure were found for progression-free survival: hazard ratios where 0.89 (0.66—1.19), 0.77 (0.59—1.01) and 0.96 (0.75—1.24), respectively.

Abstracts The prognostic significance of p53, KI67 and k-ras molecular markers was also assessed in a subset of pts (378, 377 and 124, respectively). Multivariate analyses suggest that none of these variables are prognostic for survival. During the last decade several new chemotherapeutic agents, including gemcitabine, taxanes, and vinorelbine, became available and have been extensively investigated in advanced NSCLC. When combined with cisplatin or carboplatin and compared with older regimens, these new regimens resulted in a 13% improvement in response rate and reduced toxicity, but only in a marginal increase in overall survival rate (27). Therefore it is unlikely that the use of these new regimens as adjuvant treatment in early stages of NSCLC will greatly change the efficacy outcomes. In conclusion, the ALPI trial failed to confirm the effectiveness of adjuvant MVP chemotherapy for patients with NSCLC. Subsequently to ALPI, three different randomised trials, one in the whole setting of resected NSCLC patients (IALT) and two in specific subgroups of resected patients (NCIC in stages IB and II, CALGB in stage IB), showed survival advantages for patients receiving adjuvant chemotherapy. A new meta-analysis including at least 4000 resected patients will become available in the near future to make more robust the level of evidence for adjuvant chemotherapy. Research partially funded by AIRC and CNRACRO. Adjuvant lung cancer trials using UFT based chemotherapy Nagahiro Saijo, M.D., Ph.D.1 , Harubumi Kato, M.D.2 1 National Cancer Center, Tokyo/Chiba; 2 Tokyo Medical University, Tokyo Uracil-tegaful (UFT) is a chemotherapy drug that most US oncologists do not recognize because it has not been approved in USA although limited numbers of clinical trials of the drug have been conducted against gastrointestinal tumor even in USA. UFT has widely been used in Japan in various tumor types and has been approved in many countries. Uracil-tegafur, a prodrug of 5FU, is one of oral fluorinated pyrimidine drugs that has been synthesized mainly by the pharmaceutical companies of Japan. Main purpose of oral fluorinated pyrimidine is to improve the delivery of low dose 5FU over time mimicking a continuous infusion of 5FU. The beneficial effect of tegafur is believed to derive from its slow conversion to 5FU through cytochrom P450 pathway. The released 5FU from the prodrug

Abstracts tegafur competes with uracil for catabolism by the rate-limiting enzyme dihydropyrimidine dehydrogenase. The presence of excess uracil is belived to decrease the degradation of 5FU maintaing a continuous drug level. Although widely been used in various diseases, there has been no confirmatory big randomized controlled trials. In the treatment of non-small cell lung cancer, small phase II study showed that the response rate of UFT was less than 10%. But the surgeons in Japan preferred to use it after surgery because of mild adverse events and oral administration. In a previous preliminary phase III trial of adjuvant chemotherapy after resection of NSCLC, UFT taken orally was shown to prolong survival especially in pathological stage I adenocarcinoma. Based on the data, Taiho Pharmaceutical Company organized the Japan Lung Cancer Research Group on Postsurgical Adjuvant Chemotherapy and conducted a randomized controlled trial against pathological stage I, adenocarcinoma. Patients were randomly assigned to UFT (250 mg) for 2 years or no treatment. From January 1994 through March 1997, 999 patients were enrolled. Twenty patients were found to be ineligible and were excluded from the analysis after randomization; 491 patients were assigned to receive uracil-tegafur and 488 were assigned to observation. The median duration of follow-up for surviving patients was 73 months. The difference in overall survival between the two groups was statistically significant in favor of the uracil-tegafur group (P = 0.04 by a stratified log-rank test). Grade 3 toxic effects occurred in 10 of the 482 patients (2%) who actually received uracil-tegafur. So far, six randomized trials, including the present one, have been conducted that compare surgery alone with adjuvant chemotherapy with uracil-tegafur. Among them, three trials have shown a survival benefit from treatment with uracil-tegafur. A meta-analysis of those six trials showed that adjuvant chemotherapy with uracil-tegafur improved the overall survival (hazard ratio for death, 0.77; 95% confidence interval, 0.63—0.94; P = 0.01). It is unclear whether patients with stage II or stage III disease benefit from treatment with uracil-tegafur and whether treatment for 1 year is equivalent to treatment for 2 years. Chemotherapy for malignant mesothelioma: the state of the art Hedy Lee Kindler, M.D. Section of Hematology/Oncology, University of Chicago, Chicago, IL, USA. Malignant mesothelioma was once considered “a disease unaffected by current therapeutic ma-

S15 neuvers”. Fortunately, this is no longer true. Today, cytotoxic agents with definite activity against mesothelioma achieve reproducible responses, improve symptoms, and prolong survival. Promising novel agents that target the biology of this disease are also undergoing evaluation, suggesting that progress against this disease will continue. The antifolates are the most active class of drugs for mesothelioma. The antifolate pemetrexed, which inhibits TS, DHFR, and GARFT, is the first drug approved for the treatment of this disease. Five of 13 mesothelioma patients responded in the initial phase I trial of the combination of pemetrexed with cisplatin. A phase I trial of the pemetrexed/carboplatin combination performed exclusively in mesothelioma patients achieved a 32% response rate and symptom improvement in 70%. Single-agent pemetrexed yielded a modest 14% response rate in a phase II trial; median survival was 10.7 months. Pemetrexed combined with gemcitabine achieved a modest response rate of 20%. The largest trial ever performed in patients with pleural mesothelioma was a randomized, single-blind, placebo-controlled phase III study in 456 patients that compared single-agent cisplatin to the combination of pemetrexed plus cisplatin. The response rate improved from 17% to 41% (P < 0.001), and median survival from 9.3 to 12.1 months (P = 0.020), in patients who received the combination regimen. Patients treated with pemetrexed/cisplatin also experienced greater symptomatic improvement. Another antifolate undergoing evaluation in mesothelioma is raltitrexed, which produced a 21% response rate and a median survival of 7 months in an EORTC trial. Following promising activity in a phase I trial of the raltitrexed/oxaliplatin combination, a phase II study achieved a response rate of 25%. The EORTC/NCIC recently reported the results of a phase III trial of cisplatin versus raltitrexed/cisplatin in 250 patients. Although response (14% versus 24%, P = 0.06) and survival (8.8 months versus 11.2 months, P = 0.056) were superior on the combination arm, these did not reach statistical significance. Single-agent gemcitabine produced response rates of 0%, 7% and 31% in 3 trials. Gemcitabine and cisplatin are synergistic in vitro, and active in vivo. The gemcitabine/cisplatin combination yielded a remarkable 48% response rate in a single-center phase II Australian study; the same investigators observed a 33% response rate and a median survival of 11.2 months using the identical regimen in a multi-center study. Three other groups have reported response rates of 26%, 16%, and 9% with different schedules of gemcitabine/cisplatin. Gemc-

S16 itabine given with carboplatin yields a 20% response rate, while gemcitabine plus oxaliplatin produced a response rate of 40% and a median survival of 13 months in a phase II study. Normal cells salvage adenine by cleavage of methylthioadenosine by methylthioadenosine phosphorylase (MTAP). MTAP gene deletions occur in about 66% of mesotheliomas. Tumors with MTAP gene deletions cannot make purines by the salvage pathway and must synthesize purines de novo. SDX-102 (l-alanosine), which inhibits de novo purine biosynthesis, is being evaluated in a phase II trial in patients with MTAP deficient tumors. Despite in vitro sensitivity of mesothelioma cell lines to paclitaxel, docetaxel, and irinotecan, these drugs have no single-agent activity. Although a 24% response rate and symptom improvement was observed in a British trial of vinorelbine, the addition of oxaliplatin to this regimen increased toxicity without enhancing efficacy. Phase II trials of irinotecan given in combination with cisplatin or cisplatin plus mitomycin C have produced response rates of 27% and 35%, respectively. Emerging knowledge about the biology of mesothelioma has led to the evaluation of targeted therapies. Vascular endothelial growth factor (VEGF) is an autocrine growth factor for mesothelioma; mesothelioma patients have significantly higher plasma VEGF levels than patients with any other solid tumor. Five VEGF inhibitors are being evaluated in this disease: SU5416, Bevacizumab, Thalidomide, PTK787, and BAY 43-9006. SU5416, an inhibitor of the tyrosine kinase activity of the VEGF receptor flk-1, has activity in mesothelioma, but is no longer being developed. Bevacizumab, a recombinant humanized anti-VEGF monoclonal antibody, is being evaluated in combination with gemcitabine/cisplatin in a 106 patient multi-center randomized phase II trial led by the University of Chicago. Thalidomide, which inhibits angiogenesis induced by VEGF and B-FGF, is being studied alone and with gemcitabine/cisplatin by several investigators. The Cancer and Leukemia Group B is currently evaluating PTK 787, an oral agent that inhibits both VEGF and PDGF. A CALGB trial of the VEGF and raf-1 kinase inhibitor BAY 43-9006 will commence shortly. Mesotheliomas overexpress the epidermal growth factor receptor (EGFR). In a CALGB trial of Gefitinib, EGFR over-expression was observed in 96% of patients, yet the response rate was only 2%, and median survival was 6.8 months. The SWOG trial of OSI-774 demonstrated similar inactivity. Platelet derived growth factor (PDGF), a potent mitogen for connective tissue cells, is an autocrine growth factor for mesothelioma.

Abstracts Imatinib, a highly selective inhibitor of the PDGF receptor tyrosine kinase, has been evaluated in phase II trials in Australia, The Netherlands, and Chicago; no activity has been observed. Many new cytotoxic and targeted agents are being evaluated for mesothelioma. The challenge ahead is to determine how to combine these active agents with each other and how to integrate them into multi-modality treatment regimens. CT screening for lung cancer: principles and results Claudia I. Henschke, Ph.D., M.D. Weill Medical College of Cornell University, New York, NY, USA In modern healthcare, one of the most public, most important and most frustratingly confusing topics is screening for a cancer. Mammographic screening for breast cancer has been, in the last few years, a particularly hotly disputed topic in scientific and public-policy circles, with the confusion and frustration widely reported and thus disseminated by the mass media. Most remarkably, this debacle has taken place against the backdrop of, and despite, a truly enormous amount of completed research designed to address the usefulness of such screening. To avoid a ‘mammography’ debacle for lung cancer, the fundamental principles of screening need to be presented as we have come to think of them. We will illustrate our approach by discussing Early Lung Cancer Action Project (ELCAP) and its sequels, the NY (New York-) and International (I-)ELCAP. Screening: Screening can be defined as pursuit of early diagnosis of a disease among pre-symptomatic people. While justification of such pursuit in a study of screening is that treatment of pre-symptomatic disease is more effective than when treatment is prompted by symptoms, the effectiveness of screening must ultimately be demonstrated. Pursuit of early diagnosis is defined by a regimen of screening. This regimen starts with the initial screening test and if this test is positive, a welldefined diagnostic algorithm is followed to rule-in the diagnosis of cancer. If the test result is negative or the diagnostic algorithm does not lead to a diagnosis of cancer, the person is referred to the next routinely scheduled screening as defined by the regimen. The regimen for the first, baseline screening may be different from that for the subsequent repeat screenings. Application of the screening regimen produces screen-diagnosed cases of cancer in each cycle of screening. A cycle starts with the performance of the initial, screening test and the subsequent diag-

Abstracts nostic work-up and ends before the next routinely scheduled rescreening. Any malignancy diagnosed as a result of a recommendation for biopsy is attributed the cycle during which the recommendation is made. Any case of cancer diagnosed outside the screening regimen is called an interimdiagnosed cancer and is attributed to the cycle of screening during which it is diagnosed. These screen- and interim-diagnosed cases of cancer characterize a particular regimen of screening. Such characterization can be summarized by aggregating these diagnosed cases into a frequency distribution by relevant prognostic factors (e.g., stage, size, histology). This frequency distribution we call the diagnostic distribution under the particular screening regimen. Typically, the diagnostic distribution under the baseline regimen is provided separately from diagnostic distribution under repeat screening as these repeat screening cases are pooled. These are general principles in screening for cancer, we now apply them to lung cancer screening. Screening for lung cancer: For lung cancer, a single and very small-scale experimental study, the Early Lung Cancer Action Project (ELCAP), demonstrated the great superiority of computed tomography (CT) imaging over ‘chest radiographic’ imaging in identifying cancerous ‘nodules’ in the lungs [1,2]. The study was done because screening for lung cancer was a worthwhile goal as treatment of early stage disease produced a distinctly higher cure rate than when it is diagnosed in late stage (70% versus 10%). Further, the introduction of helical (spiral) CT imaging in the early 1990s provided a more promising test for detection of smaller nodules in the lungs than provided by traditional chest radiography as images of the chest could be obtained in less than 20 s at a low dose of radiation without the use of intravenous contrast material. In the ELCAP, the regimen of screening started with the initial, low-dose CT [1]. In that initial study, we defined a positive result of baseline screening to be 1-6 non-calcified nodules. If none were identified, the result was negative and the person was referred to the first annual repeat screening 1 year later. The baseline screening of initial 1000, initially produced 27 screen-diagnosed and 2 interim-diagnosed endobronchial lung cancers. The diagnostic distribution under baseline screening of these screen-diagnosed cases by stage and size was given in the Lancet publication in 1999. The definition of positive result of annual repeat screening was defined as any growing non-calcified nodule, regardless of size [2]. Among the 1184 instances of annual repeat screening of the 1000, 7 were diagnosed with lung cancer; no interim-

S17 diagnosed cancers were identified in this cycle of repeat of screening. Important components of the ELCAP diagnostic algorithm was assessment of nodule growth and CT-guided fine-needle aspiration, particularly for nodules 15 mm or less. Growth was assessed by comparison of the CT images obtained at different times, both visually and by image-analytic techniques [3—6]. If growth was consistent with malignancy, fine-needle aspiration was recommended. Both growth and fine-needle aspiration are important components of a program of CT screening for lung cancer as they serve to limit unnecessary surgery. Since the initial results of ELCAP were published, helical CT scanning has rapidly advanced from the single-slice scanner to 4-, 8-, 16- and soon even more-slice scanners. Using these multi-slice scanners, images of less than 1 mm slice thickness can be obtained in a single breathhold as compared to 10 mm slice thickness using single-slice scanners. As resolution was markedly improved, it is not surprisingly that many more nodules are being detected, most of them well below 5.0 mm. Accordingly, the definition of positive result on baseline screening required updating. A new type of nodule, the sub-solid one, was identified on CT screening tests [7]. Sub-solid nodules had not been recognized on chest radiographs nor was their importance understood initially in CT screening. Upon review of the ELCAP results, we found that sub-solid nodules consisting of either part-solid or non-solid ones, had a higher malignancy rate than solid ones and that part-solid nodules had about a three-fold higher malignancy rate. As a result of these new findings and the advances of CT scanning, the definition of a positive result of screening has been changed from 1 to 6 non-calcified nodules to one or more non-calcified solid or part-solid nodules 5.0 mm or larger or a nonsolid nodule 10.0 mm or larger (without solid or part-solid nodules 5.0 mm or greater) [8]. Using this updated definition, the percentage of positive result on the baseline, low-dose screening is reduced to around 10% without any increase in the false negative rate. The definition for positive results on annual repeat screening has remained unchanged. Definition of the particulars of the screening regimen starting with the initial test to the diagnosis of lung cancer is critical to understanding the diagnostic distribution achieved under the particular regimen. Another regimen of screening will produce a different diagnostic distribution and thus, when comparing the results of studies, the differences in regimens must be understood. Also, it should be realized that the diagnostic distribution can be

S18 obtained to any desired accuracy by performing a baseline and single repeat screening with sufficient numbers of screenees. Comparison of two screening regimens or any component of the regimen, such as different screening tests, requires determination of their respective diagnostic distributions. The ELCAP provides such an example as baseline screening was done using both chest radiography and CT scan on everyone. The two resulting diagnostic distributions were provided by stage and size in the initial publication in 1999 [1]. Comparison of these two distributions demonstrated the great superiority of computed tomography (CT) imaging over chest radiography in identifying cancerous ‘nodules’ in the lungs. At present, it is generally accepted that lowdose CT screening leads to early diagnosis of lung cancer in a high percentage of the cases. Based on prior evidence, annual CT screening provides for detecting the disease at earlier and presumably more commonly curable stages in a cost-effective manner. Ongoing projects: The ELCAP report, in turn, inadvertently led to considerable public and professional interest in the practice of CT-based screening for lung cancer [9]; and within Cornell it led to two carefully-considered initiatives: the planning and fund-raising for a project experimentally to screen 10,000 high-risk persons in what got to be called the New York ELCAP (NY-ELCAP) [10], and the International Conferences on Screening for Lung Cancer [11]. The latter initiative also was an outgrowth of the Cornell team’s already extensive role in helping other investigator groups in various institutions initiate research projects patterned after the original ELCAP, including sharing with them its web-based management and data-recording system and its associated teaching files. Integral to this international collaboration is pooling of the data, which represents the final element in what is now being launched as the International-ELCAP (I-ELCAP) [12], simultaneously with the NY-ELCAP and with a shared set of not only principles but protocol as well. The I-ELCAP protocol represents an update of that in the original ELCAP, as is to be expected in this novel and rapidly evolving area of medical technology and its requisite evaluation toward the knowledge-base of future practice. At any given time, the I-ELCAP protocol focuses on the then-most-promising regimen of early (pre-symptomatic) diagnosis of lung cancer; and in respect to any such regimen, the first-order aim is to determine how early the diagnoses are achieved [8]. The diagnostic subclassification is fur-

Abstracts ther refined by other prognostic indicators, notably cell type and, for the earliest diagnoses, CTbased measures of the tumor’s rate of growth. Focus is on diagnoses in the practice-relevant sense of intervention-guiding, pre-surgical diagnosis. For each of the diagnostic-prognostic categories, the broad aim is to determine their respective degrees of significance. This involves determining their associated case-fatality rates without and with intervention (in the absence of competing causes of death), together with the respective timings of fatal outcomes. Implied by these are the category-specific proportions of overdiagnosed cases, curability rates for progressive cases, and the time lags to deaths preventable by early intervention. With sufficiently detailed diagnosticprognostic categories, these parameters presumably are general over various regimens of early diagnosis. The category-specific, fundamental results under these two aims (frequency and significance) imply the overall case-fatality and curability rates associated with the regimen of early diagnosis, and also that specific to screen-detected cases—–both of these for genuine, progressive cases of lung cancer. They also imply the timing of the deaths prevented by screening-associated early intervention. For this reason, we can an expert panel review all malignancies according to a detailed protocol [13,14].

References [1] Henschke CI, McCauley DI, Yankelevitz DF, Naidich DP, McGuinness G, Miettinen OS, et al. Early lung cancer action project: overall design and findings from baseline screening. Lancet 1999;354:99—105. [2] Henschke CI, Naidich DP, Yankelevitz DF, McCauley DI, McGuinness G, Smith JP, et al. Early lung cancer action project: initial findings on repeat screening. Cancer 2001;92:153—9. [3] Yankelevitz DF, Gupta R, Zhao B, Henschke CI. Repeat CT scanning for evaluation of small pulmonary nodules: preliminary results. Radiology 1999;212:561—6. [4] Yankelevitz DF, Reeves A, Kostis W, Zhao B, Henschke CI. Determination of malignancy in small pulmonary nodules based on volumetrically determined growth rates: preliminary results. Radiology 2000;217(1):251—6. [5] Reeves A, Chan A, Yankelevitz D, Henschke C, Kostis WJ. On measuring the change in size of pulmonary nodules. IEEE Transaction on Medical Imaging, in press. [6] Kostis WJ, Reeves AP, Yankelevitz DF, Henschke CI. Threedimensional segmentation and growth rate estimation of small pulmonary nodules in helical CT images. IEEE Tran. Med Imaging 2003;22:1259—74. [7] Henschke CI, Yankelevitz DF, Mirtcheva R, McGuinness G, McCauley D, Miettinen OS. CT screening for lung cancer: frequency and significance of part-solid and nonsolid nodules. AJR 2002;178:1053—7.

Abstracts [8] I-ELCAP protocol, website: http://ICScreen.med.cornell. edu. [9] New York Times, front page, July 9, 1999. [10] New York Early Lung Cancer Action Program, website: http://www.NYELCAP.org. [11] International Collaboration to Screen for Lung Cancer. In: Proceedings of the First, Second, Third, Fourth, Fifth, and Sixth International Conference on Screening for Lung Cancer. New York, NY, website: http://ICScreen.med. cornell.edu. [12] International Early Lung Cancer Action Program, website: http://www.IELCAP.org. [13] Vazquez M, Flieder D, Travis W, Carter D, Yankelevitz D, Miettinen OSM, et al. Early lung cancer action project pathology protocol. Lung Cancer 2003;39:231—2. [14] Vazquez M, Flieder D, Travis W, Carter D, Yankelevitz D, Miettinen OSM, et al. Early lung cancer action project pathology protocol, website: http://ICScreen.med.cornell.edu.

Head and neck cancer trials combining EGFR inhibition with radiation Paul M. Harari, M.D. University of Wisconsin Comprehensive Cancer Center, Madison, WI, USA Over the last several decades, dominant clinical research themes in H&N cancer have largely involved intensification of radiation and chemotherapy. This research effort has commonly taken the form of either altered radiation fractionation or chemoradiation. During recent years in particular, concurrent chemoradiation has assumed a leading role in clinical trials and in global H&N cancer practice. Modest gains in locoregional disease control and overall survival have generally been achieved at the expense of higher acute and sometimes chronic toxicities. The prospect that new molecular therapies might complement or augment conventional cancer treatment modalities is under intense clinical investigation in virtually all major tumor types. The first randomized clinical trial (phase III) to identify a survival advantage when combining an EGFR inhibitor with conventional cancer therapy has emerged in H&N cancer (ASCO 2004). This study randomized 424 stage III/IV H&N cancer patients to curativeintent radiation alone ± weekly cetuximab (antiEGFR mAb). There was a statistically significant improvement (log-rank P = 0.02) in locoregional disease control (8% at 2 years) and overall survival (13% at 3 years) favoring the experimental arm. Cetuximab treatment was generally well tolerated with the most common adverse event being that of skin reaction. There are several potential implications of this international H&N cancer trial. First, the results identify the capacity of a molecular targeted growth inhibitor (cetuximab in this case) to aug-

S19 ment survival outcome in a curative cohort of patients treated with radiation alone. This finding suggests that other epithelial cancers treated with dominant radiation approaches may warrant clinical investigation regarding the addition of EGFR inhibitors. Second, the results suggest that cetuximab (and possibly other EGFR inhibitors) warrant clinical evaluation regarding their capacity to enhance outcome when added to chemoradiation regimens in advanced H&N cancer. Finally, the similar survival gains observed with cetuximab/radiation to those observed with chemoradiation (compared against radiation alone) suggest the potential for cetuximab to substitute for cisplatin in advanced H&N cancer therapy; a concept that also warrants systematic clinical evaluation. Nonsurgical treatment for advanced laryngeal cancer Merrill S. Kies, Jan Lewin University of Texas MD Anderson Cancer Center, Houston, TX, USA Although modern conservation surgical techniques are applicable for selected patients with squamous cell carcinoma of the larynx, systemic chemotherapy and radiotherapy have more often provided nonsurgical alternatives to the treatment of advanced laryngeal cancers. Moreover, a sequence of randomized trials has demonstrated that the concomitant administration of chemotherapy and radiotherapy leads to improved local—regional disease control and overall survival compared to radiotherapy administered as a single modality, in patients with locally advanced squamous cell carcinomas of the head and neck [1—6]. Induction chemotherapy with cisplatin and fluorouracil has been recognized as highly active, with clinical partial and complete responses observed in 80%—90% of previously untreated patients [7—9]. It was postulated that a substantial response to initial treatment with chemotherapy would lead to an improvement of therapeutic efficacy for surgery or radiotherapy. This led to the Department of Veterans Affairs Laryngeal Cancer Study [10], in which 332 patients with stage III or IV squamous cell carcinoma of the larynx were randomized to receive either induction chemotherapy consisting of cisplatin and fluorouracil followed by radiotherapy or surgery and postoperative radiotherapy. Patients who experienced no tumor response to chemotherapy or those who had locally persistent or recurrent cancer underwent salvage laryngectomy. Two-year survival for both treatment groups was 68%, and 41% of patients randomly assigned to the experimental

S20 arm were alive with a functional larynx at 2 years. Thus, the efficacy of chemotherapy followed by radiotherapy (with surgical salvage) was similar to that of surgery followed by radiotherapy and established organ preservation as a realistic goal of nonsurgical treatment administered with curative intent. Lefebvre et al. [11] later reported similar outcomes in a European trial involving patients with cancers of the hypopharynx. In the Veterans Affairs study [10], there were observed trends in patterns of tumor relapse, with 20% of patients in the chemotherapy arm having locoregional recurrence versus 7% in the surgery arm. Distant disease recurrence was more likely in the surgical arm, affecting 17% of patients versus 11% in the chemotherapy/radiotherapy arm. Salvage laryngectomy was required more often in patients with glottic cancers than in those with supraglottic primary sites (43% versus 31%), in patients with fixed vocal cords than in those with mobile vocal cords (41% versus 29%), and in patients with gross invasion of thyroid cartilage than in patients with (41% versus 35%). Notably, salvage laryngectomy was required in 56% of patients with T4 cancers compared with 29% of patients with smaller primary tumors (P = 0.001). The Veterans Affairs larynx study prompted further investigations of chemotherapy and radiotherapy in the treatment of larynx cancer using the sequential administration of induction chemotherapy, consisting of cisplatin and fluorouracil, followed by radiotherapy a the control arm. This was compared with concomitant cisplatin and radiotherapy and radiotherapy administered as a single treatment modality [12,13]. For all groups, totaling 547 patients, surgical salvage was reserved for those patients with persistent or locally recurrent disease. Survival did not see to be affected by treatment assignment. At a median follow-up of 3.8 years, patients randomly assigned to concomitant cisplatin and radiotherapy achieved a higher rate of laryngeal preservation, 84% versus 72% in patients receiving sequential chemoradiotherapy (P = 0.005) or radiotherapy alone 67% (P < 0.001). At 2 years, 80% of patients on the concomitant chemoradiotherapy arm achieved local control versus 64% of patients receiving sequential chemotherapy and then radiation and 58% of patients treated with radiotherapy alone. As expected, acute toxic effects of treatment were greatest in the patients who received radiotherapy with concomitant cisplatin. These trials indicate that for patients with intermediate stage squamous cell carcinoma of the larynx, a combined treatment program with the objectives of tumor eradication and laryngeal preservation is appropriate. It is also important

Abstracts to recognize that patients with locally advanced, destructive primary laryngeal cancers were not included in the recent multi-group trial. These patients may require total laryngectomy for optimal tumor control and to preserve function. Impact of nonsurgical treatment of advanced laryngeal cancer on speech and swallowing function As just described, combinations of chemotherapy and radiation for laryngeal preservation are well-documented alternatives to ablative surgery for advanced stage disease, but there are few studies that have objectively addressed the issue of preservation of function. Radiation produces tissue changed that can result in immediate and long-term alterations in speech and swallowing. The magnitude of the problem will frequently depend on the administered dose and duration of treatment, the extent of the treatment field, and whether concomitant chemotherapy was used. In general, radiotherapy has a greater impact on swallowing than it does on speech [5,14,15], and the addition of chemotherapy can exacerbate these effects [15]. The adverse impact of radiation may equal or exceed that associated with surgery because the sequelae may increase in severity years after the completion of treatment. Some studies comparing swallowing function in radiated patients at 6 months and 10 years post-treatment have shown similar abnormalities in the pharyngeal swallow [16]. Contributory factors to this poor swallowing ability include tissue changes, fibrosis, and possible alterations in sensory awareness. In addition to restricting laryngeal movements, radiation-induced fibrosis can reduce the range of motion of the tongue and jaw and can diminish pharyngeal wall motion when the larynx is in the radiation field. In a recent study of 10 patients treated with chemoradiotherapy, results demonstrated persistent and severe swallowing dysfunction beyond 1 year after treatment completion [15]. Radiation to the larynx usually results in swallowing problems related to pharyngeal transport. Again, restricted laryngeal motion, impaired true vocal fold closure, and aspiration are common complications in patients who have been treated with radiation. Similarly, dysphonia is often present, but its impact on daily oral communication does not seem to be as significant when compared with the deficits that occur in swallowing [17]. To counteract the deleterious effects of radiation and chemoradiotherapy, rehabilitative options exist and are best administered by a qualified speech pathologist. These interventions include exercise protocols to strengthen and maintain

Abstracts range of motion, precision, muscle elasticity and mobility and to teach compensatory strategies that should be started early, preferably at the initiation of treatment and continued throughout the course of treatment. These specific exercises have been shown to improve pharyngeal mobility during the swallow and result in enhanced pharyngeal clearance, airway protection, and elimination of aspiration in the majority of cases [18,19]. The long-term use of this exercise regimen may be necessary to counter the long-term adverse effects in patients who have received irradiation. Prospectus: We also will review investigational strategies chemotherapy and radiation as primary definitive treatment for larynx cancer. Ongoing trials will be discussed. Furthermore, we will consider the potential integration of newer “targeting” compounds for the initial treatment of cancers of the larynx.

References [1] Merlano M, Vitale V, Rosso R, et al. Treatment of advanced squamous cell carcinoma of head and neck with radio- and chemotherapy alternatively. N Engl J Med 1992;327:1115—21. [2] Brizel DM, Albers ME, Fisher SR, et al. Hyperfractionated irradiation with or without concurrent chemotherapy for locally advanced head and neck cancer. N Engl J Med 1998;338:1798—804. [3] Wendt TG, Grabenbauer GG, R¨ odel CM, et al. Simultaneous radiochemotherapy versus radiotherapy alone in advanced head and neck cancer: a randomized multicenter study. J Clin Oncol 1998;16:1318—24. [4] Calais G, Alfonsi M, Bardet E, et al. Randomized trial of radiation therapy versus concomitant chemotherapy and radiation therapy for advanced-stage oropharynx carcinoma. J Natl Cancer Inst 1999;91:2081—6. [5] Jeremic B, Shibamoto Y, Milicic B, et al. Hyperfractionated radiation therapy with or without concurrent low-dose daily cisplatin in locally advanced squamous cell carcinoma of the head and neck: a prospective randomized trial. J Clin Oncol 2000;18:1458—64. [6] Adelstein DJ, Li Y, Adams GL, et al. An intergroup phase III comparison of standard radiation therapy and two schedules of concurrent chemo radiotherapy in patients with unresectable squamous cell head and neck cancer. J Clin Oncol 2003;21:92—9. [7] Ensley JF, Jacobs JR, Weaver A, et al. Correlation between response to cisplatinum-combination chemotherapy and subsequent radiotherapy in previously untreated patients with advanced squamous cell cancers of the head and neck. Cancer 1984;54:811—14. [8] Kies MS, Gord LI, Hauch WW, et al. Analysis of complete responders after initial treatment with chemotherapy in head and neck cancer. Otolaryngol Head Neck Surg 1985;93:199—205. [9] Ervin TJ, Clark JR, Weichselbaum RR, et al. An analysis of induction and adjuvant chemotherapy in the multidisciplinary treatment of squamous-cell carcinoma of the head and neck. J Clin Oncol 1987;5:10—20.

S21 [10] The Department of Veterans Affairs Laryngeal Cancer Study Group. Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer. N Engl J Med 1991;324:1685— 90. [11] Lefebvre JL, Chevalier C, Luboinski B, et al. Larynx preservation in pyriform sinus cancer: preliminary results of a European Organization for Research and Treatment of Cancer phase III trial. J Natl Cancer Inst 1996;88:890—9. [12] Forastiere AA, Goepfert H, Maor M, et al. Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 2003;349:2091—8. [13] Weber RS, Berkey BA, Forastiere A, et al. Outcome of salvage total laryngectomy following organ preservation therapy: The Radiation Therapy Oncology Group trial 91—11. Arch Otolaryngol Head Neck Surg 2003;129(1):44—9. [14] Pauloski BR, Rademaker AW, Logemann JA, et al. Speech and swallowing in irradiated and nonirradiated postsurgical oral cancer patients. Otolaryngol Head Neck Surg 1998;118(5):616—24. [15] Smith RV, Tamar K, Beitler JJ, et al. Long-term swallowing problems after organ preservation therapy with concomitant radiation therapy and intravenous hydroxyurea. Arch Otolaryngol Head Surg 2000;126:384—9. [16] Kendal KA, McKenzie SW, Leonard RJ, et al. Timing of swallowing events after single-modality treatment of head and neck carcinomas with radiotherapy. Ann Otol Rhinol Laryngol 2000;109:767—75. [17] Carrara-de Angellis E, Feher O, Barros APB, et al. Voice and swallowing in patients enrolled in a larynx preservation trial. Arch Otolaryngol Head Neck Surg 2003;129:733—8. [18] Lazarus CL, Logemann JA, Pauloski BR, et al. Swallowing disorders in head and neck cancer patients treated with radiotherapy and adjuvant chemotherapy. Laryngoscope 1996;106:1157. [19] Lazarus CL. Effects of radiation therapy and voluntary maneuvers on swallowing function in head and neck cancer patients. Clin Commun Disord 1993;3:11.

Advances in systemic treatment and monitoring approaches in NPC Anthony T.C. Chan Chinese University of Hong Kong, Hong Kong, PRC With improvement in local control achieved by modern tumor imaging and radiotherapy techniques, distant metastasis has become the leading cause of NPC failure. Attempts to improve the treatment outcome have therefore focused on incorporating optimal systemic therapy into the primary radiotherapy treatment. Two randomized studies of adjuvant chemotherapy were negative and six randomized studies of neoadjuvant chemotherapy also failed to demonstrate improvement in overall survival (OS), although two demonstrated improvement in local control and progression free survival (PFS). Two published randomized studies of concurrent chemoradiotherapy from US Intergroup (with adjuvant chemotherapy) and Taiwan (without adjuvant chemotherapy) demonstrated significant improvement in OS and PFS. The updated results of a multicenter ran-

S22 domized study from Hong Kong using concurrent weekly cisplatin 40 mg/m2 and radiotherapy also demonstrated significant OS improvement. Preliminary data on the use of neoadjuvant chemotherapy using newer agents including taxol, taxotere and gemcitabine followed by concurrent chemoradiation have been encouraging and are being further investigated in the randomized setting. In the metastatic setting, first-line platinumbased doublets have consistently achieved 50—80% response rates with those having lung metastases alone representing a distinctive group of significantly better OS. Ongoing studies include combination of newer agents like gemcitabine and oxaliplatin aiming to reduce the toxicity profile. The epidermal growth factor receptor (EGFR) is over-expressed in NPC and in vitro data supports the combination of C225 with chemotherapy. A multicenter phase II study of C225 and carboplatin in platinum-failure patients achieved a response rate of 11.7% and stable disease in 48.3% of patients. A phase II study of Iressa in second-line metastatic NPC is currently ongoing. Endemic NPC is universally associated with the Epstein-Barr virus (EBV), providing a unique opportunity for molecular-targeted therapy as well as monitoring. Our center has pioneered the use of quantitative cell-free EBV DNA in plasma of NPC patients as a highly sensitive and specific marker, as well as a powerful tool in predicting clinical outcome and disease monitoring. DNA methylation of EBV promoter regions result in silencing of viral gene expression, allowing tumors to escape from immune recognition. A pilot study recently demonstrated that azacitidine can produce demethylation of specific DNA sites in tumors. This has opened up a novel therapeutic approach targeted at reactivating the expression of epigenetically inactivated genes. Tongue cancer 2004 Jeffrey N. Myers M.D., Ph.D. Department of Head and Neck Surgery, University of Texas, M.D. Anderson Cancer Center, TX, USA Squamous cell carcinoma of the oral tongue (SCCOT) is one of the most common malignant tumors of the upper aerodigestive tract (UADT), accounting for 6900 new cancer cases and 1700 deaths from cancer each year. For unclear reasons, over the past several decades, the incidence of SCCOT is rising in individuals without the traditional risk factors of advanced age, and excessive tobacco and/or alcohol use. Given the tongue’s critical functions in speech and swallowing, these tumors and their treatment can result in significant

Abstracts morbidity for patients stricken with this disease. While organ preservation strategies for cancers arising in other head and neck sites such as the oropharynx, hypopharynx, and larynx have made significant in-roads in providing local—regional control and survival rates comparable to those achieved with surgery and adjuvant radiotherapy; tongue cancer remains a tumor that is treated primarily with surgery. When matched by stage to tumors at other UADT sites, regional control and survival rates in SCCOT are less favorable. The major clinico-pathologic predictors of failure are advanced clinical T and N stage, poorly differentiated histology, peri-neural invasion, the number of lymph nodes involved with tumor, and the presence of extra-capsular spread of tumor outside the lymph node. The relatively high rates of regional recurrence, distant metastases, and death from disease indicate that (1) intensification of therapy is warranted and that (2) enhanced understanding of the biology of this tumor is required to develop new therapeutic targets. Intensification of therapy through the use of biologic agents and chemotherapy in addition to post-operative adjuvant radiotherapy for patients with high risk pathology is currently under investigation. Furthermore, pre-clinical advances in the development of an orthotopic tumor model which recapitulates the pathways of tumor growth and spread has provided a unique opportunity to unravel the mechanisms of tongue tumor growth, survival, and spread and to evaluate novel targeted therapies. Developing smoking cessation programs for head and neck and lung cancer patients John A. Ridge, Robert A. Schnoll Fox Chase Cancer Center, Department of Surgical Oncology, Philadelphia, PA, USA. Background: Continued smoking after being diagnosed with a cancer has important adverse consequences. It reduces survival time, increases the risk of developing a recurrence or second primary cancer, and limits the efficacy of treatment while increasing complications. Nonetheless, many patients with head and neck and lung cancers continue to smoke. As a result, there is an increasing clinical and academic interest in smoking cessation. When approached, some patients are willing to participate in smoking cessation programs and associated trials, whereas others are not. Differences between these groups of patients should be addressed in order to facilitate enrollment and to tailor interventions. Methods: More than 580 patients presenting to the head and neck and lung cancer multidisciplinary

Abstracts clinics have been screened for eligibility to pursue a no-cost smoking cessation program. More than 230 proved to be candidates. Although eligibility criteria were less stringent, all participants reported smoking within the prior 30 days. Characteristics of those both willing and unwilling to engage in the program (which involved education, counseling, skills-training to quit and to avoid relapse, and nicotine replacement therapy) were evaluated. Results: More than half of eligible patients declined to participate in the cessation program. Those patients were significantly more likely to: have head and neck cancer (rather than lung cancer); have fewer physical symptoms; show no intention to quit smoking (or report less readiness to quit); and smoke fewer cigarettes. Nearly half cited a desire to quit without formal help as a reason not to enter the smoking cessation program. Twentytwo percent expressed no interest in quitting. Only those patients entering the program were available for evaluation of their psychological profiles and motivation to quit smoking. Nearly 80% had tried to stop “cold turkey.” Four percent exhibited high levels of fatalism and only 10% believed that there were profound disadvantages to quitting, while more than half acknowledged advantages. Lack of confidence in one’s ability to quit and significant depressive symptoms were found in more than 40% of those agreeing to enroll. Willingness to enter, but with low motivation to quit was associated with: a longer interval since diagnosis and approach; more cigarettes smoked per day; lower perception of smoking’s risks; and a lack of self-confidence surrounding the ability to quit. Conclusions: There are differences between those patients who agreed to enter the smoking cessation program and those who declined to do so. Efforts to increase enrollment and to diminish failure and relapse should address both groups, targeting: motivation, risk perception, benefits of quitting, and depression. Individual patient data meta-analyses in head and neck carcinomas: what have we learnt? J. Bourhis, B. Baujat, J.P. Pignon, on behalf of the MACH-NC, MARCH Collaborative Groups Institut Gustave Roussy, Villejuif, France The meta-analyses reported here are based on updated individual patients data, the gold standard method for meta-analysis of randomized trials. For head and neck squamous cell carcinomas, two data bases of randomized trials have been generated: (1) the MACH-NC data base (17858 patients), evaluating the question of adding chemotherapy (CT) to local treatment and (2) the MARCH data base, eval-

S23 uating the question of altered fractionated radiotherapy (RT) (6515 patients). A third meta-analysis was performed on larynx preservation. For the MACH-NC, the main results were that the benefit associated with the use of CT depended on the timing of CT, concomitant RT—CT being more effective than adjuvant or neo-adjuvant CT. The overall improvement in survival at 5 years in favor of adding CT concomitantly to RT was 8%, and more pronounced when CDDP alone was used (11%). The effect of poly or mono chemotherapy were not found to be statistically different, when given concomitantly to RT. The benefit associated with the use of concomitant CT was decreasing significantly with age, and more pronounced in younger patients. The effect of concomitant CT was found relatively unchanged, whether RT was conventional, altered fractionnated RT or adjuvant RT after surgery. For the MARCH, the main results were that the benefit associated with the use of altered fractionnated RT depended on the type of RT, hyperfractionnated RT being more effective than other types of altered fractionated RT. The overall improvement in survival at 5 years in favor of altered fractionated RT, as compared to standard RT alone was 3%, 6% and nearly 10% regarding respectively survival, local—regional control and local control. The benefit associated with the use of altered fractionated RT was decreasing significantly with age, and more pronounced in younger patients. In conclusion, both the addition of CT and altered fractionated RT; significantly improved survival. The evolution of sequential therapy: combining induction chemotherapy; chemoradiotherapy, and surgery in the curative treatment of advanced head and neck cancer Marshall R. Posner, Roy B. Tishler, Charles M. Norris, Lori Wirth, Robert I. Haddad Dana-Farber Cancer Institute, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA Induction chemotherapy, defined as chemotherapy prior to definitive surgery and/or radiotherapy was first proposed for squamous cell cancer of the head and neck (SCCHN) in the 1980s as a result of evidence that combination chemotherapy was highly effective in shrinking advanced tumors. Early randomized trials failed to demonstrate an effect on survival because of substantial flaws in treatment plans trial design, patient selection and statistical power. These results led to discouragement and a sense that induction chemotherapy was ineffective

S24 in improving survival. Similar problems effected chemoradiotherapy trials. Meta-analysis of a large number of randomized trials demonstrates that chemotherapy improves survival compared to standard treatment. The effect of chemoradiotherapy on 5-year survival is highly significant, while a significant positive effect of induction chemotherapy is limited to the subset of trials that delivered platinum plus five-fluorouracil, based on the original Wayne State regimen (PF). Large, well-designed and executed, randomized trials have shown that PF based induction chemotherapy followed by surgery and/or radiotherapy results in significantly improved survival compared to standard therapy alone. Large, platinum-based chemoradiotherapy trials have also shown improved survival compared to radiotherapy alone in patients with advanced disease. Recently, two large randomized trials have demonstrated that induction chemotherapy with a taxane added to PF (TPF) resulted in improved survival compared to PF. In addition, TPF is associated with less toxicity then PF. This is the first demonstration of a treatment that improves on PF induction chemotherapy. TPF can be considered the standard of care when induction chemotherapy is part of the treatment plan. TPF represents the platform for the addition of molecularly targeted agents to induction chemotherapy. Sequential therapy, the combination of induction chemotherapy with chemoradiotherapy makes sound biological sense. While improving local and regional control, chemoradiotherapy has not impacted on the control of distant disease. In contrast, many patients who receive induction chemotherapy and standard radiotherapy are controlled distantly, but fail locally. Based on these results and exploratory studies that indicated that chemoradiotherapy could salvage failures to induction chemotherapy, sequential therapy treatment plans, combining induction chemotherapy and chemoradiotherapy have been explored to address weakness in both treatment paradigms. Sequential therapy appears very promising. There are now randomized trials comparing sequential therapy to chemoradiotherapy either ongoing or planned. These trials will help determine if chemoradiotherapy or sequential therapy represents the next standard of care for patients with locally advanced SCCHN.

Abstracts Antitumor efficacy of ZD6474, a VEGFR tyrosine kinase signaling inhibitor with additional activity against EGFR tyrosine kinase, alone and in combination with radiotherapy in human head and neck cancer David Raben, Daniel L. Gustafson, Barb Frederick, Andrea L. Merz, Joseph A. Zirrolli University of Colorado Health Sciences Center, Denver, CO, USA Background: The response rates to epidermal growth factor receptor (EGFR) associated tyrosine kinase inhibitors (TKIs) appear to be approximately 10−15%. Reasons for this level of response may include varying downstream signaling pathways despite similar histologies, mutations in specific parts of the TK sequence that effect response, and overriding proangiogenic signaling by unknown mechanisms. In this regard, the inhibition of tumor angiogenesis should have wide applicability in all solid malignancies. ZD6474 is an oral 4-anilinoquinazoline that inhibits TK activity associated with VEGFR-2 and EGFR. Previous studies of ZD6474 in combination with certain cytotoxic agents (paclitaxel, docetaxel or gemcitabine) have shown at least additive tumor growth inhibition. Additionally, ZD6474 appears to enhance the effects of radiotherapy (RT) of the tumor and throughout the surrounding endothelium. The purpose of our investigations is to assess the activity of ZD6474 alone and in combination with RT against gefitinib-sensitive and -insensitive HNSCC cells in vitro and in vivo. An emphasis of our investigations is to evaluate the optimal methods of translating therapeutic scheduling of TKIs with radiation. Materials and methods: In vitro MTT assays were performed with ZD6474 alone and in combination with gefitinib or radiation. Western blot analysis was also performed to evaluate the effects of ZD6474 versus gefitinib on phosphorylated proteins associated with EGFR signaling. Subsequently, nude mice were implanted with human gefitinibsensitive, EGFR+ , UMSCC2 tumors. Mice bearing tumors (average size 300 mm3 ) were treated with ZD6474 alone (30 mg/kg per day), RT alone (2 Gy × 3 Gy per week for 2 weeks) and combinations of ZD6474 and RT including concomitant therapy, ZD6474 followed by RT, and RT followed by ZD6474. Tumor and plasma samples were also collected during ZD6474 therapy and drug levels measured. Additional animal experiments evaluated the effects of ZD6474 in EGFR− and gefitinib-insensitive tumors.

Abstracts

S25

Group 1

Group 2

Group 3

Group 4

Group 5

Group 6

UMSCC2 tumor growth delay Tumor doubling time (days) (mean ± S.D.)

14.3 ± 6.0

18.9 ± 5.6

31.1 ± 13.2∗

35.7 ± 6.8∗∗∗ 30.7 + 11.2∗

22.8 ± 7.1

Median tumor doubling time (days)

11.0

17.0

36.0

37.0

22.5

Fraction of animals with tumors that did not double in size by day 47

0/9

0/9

1/8

4/10

29.5 3/9

5/9

Levels of significance is P < 0.05 as determined by ANOVA analysis with Tukey’s pairwise multiple comparison procedures. ∗ Significantly different from group 1. ∗∗ Significantly different from group 2.

Results: ZD6474 growth inhibition was examined in a panel of HNC cell lines in vitro. In some of the lines, greater activity was observed compared with gefitinib. In ZD6474-sensitive HNC cell lines, enhanced radiosensitivity was seen; however, the effects were dose- and cell line-dependent. The results of the initial studies in vivo are shown below. Group 1 is untreated; Group 2 is RT alone on days 1−14; Group 3 is ZD6474 alone on days 1−14; Group 4 is ZD6474 and RT combined on days 1−14; Group 5 is RT on days 1−14 and ZD6474 on days 15−28; and Group 6 is ZD6474 on days 1−14 and RT on days 15−28. Plasma and tumor levels of ZD6474 were measured 6 h after dosing for 5 consecutive days in a subset of animals treated with ZD6474 alone. ZD6474 plasma levels were determined to be 3.64 ± 1.12 ␮M and tumor levels were measured as 0.073 ± 0.024 ␮mol/g. Conclusions: ZD6474 is an attractive agent for preclinical and clinical evaluation based on its oral bioavailability, activity against both VEGFR-2 and EGFR signaling, inhibition of angiogenesis, dosedependent inhibition of tumor growth in a variety of tumor models, and its radiosensitization effects. The results from our preclinical studies in HNC suggest a therapeutic benefit for combined RT and ZD6474 therapy, both sequentially and concurrently. This is most clearly illustrated by the increased fraction of animals whose tumors did not double in size in the groups treated with combination therapy when compared with ZD6474 alone. Furthermore, the pharmacokinetic results from our work indicate that the plasma levels of drug obtained in our preclinical studies are within the range of that seen in patients treated with ZD6474 in Phase I studies. Data will also be presented from both EGFR− and gefitinibinsensitive HNC cell lines treated with ZD6464. Future studies are aimed at elucidating the mechanism by which ZD6474 enhances radiotherapy, and determining if optimal RT schedules are cell linedependent.

Does EGFR expression level predict for tumor response to therapy Kian Ang The University of Texas, M.D. Anderson Cancer Center, Houston, TX, USA Targeting specific biological pathways in tumor development has been heralded as a promising approach to the treatment of cancer. Familiar to most investigators are the studies done with epidermal growth factor receptor (EGFR) antagonists. The rationale behind the use of EGFR inhibitors was based on the fact that EGFR activation stimulates many important signaling pathways associated with cancer development and progression, and importantly, resistance to radiation. Since EGFR overexpression portends for a worse outcome in patients with advanced HNC, selective targeting of this signaling pathway has gained attention and emerging results of clinical trials on patients with metastatic (recurrent) lung, head and neck (HN), and colorectal cancers are indeed encouraging. Recently reported data (ASCO 2004) on the combination of radiation with C225 for the primary treatment of patients with locally advanced HN carcinoma are impressive, i.e., showing improvement of local—regional control and survival rates relative to radiation alone. Although the collective clinical outcome data are exciting, the data on the correlation between the magnitude of tumor EGFR expression and response to therapy are ambiguous. Confounding factors causing part of the confusion include lack of standardization of tissue acquisition and storage and immunological assay methodology (e.g., type of antibody, quantitative scoring) and the variations in the therapy regimens tested. Collectively, there appear to be no or minimal correlation between EGFR expression level and tumor response to EGFR antagonists given as single agents or to cytotoxic agents with or without EGFR antagonists, though it is prudent to confirm this impression through well designed prospective studies. The data on the

S26 correlation between EGFR expression level and tumor response to radiotherapy is, however, quite strong and have undergone stringent validation. The available data will be summarized in this paper. Molecular targets for lung cancer therapy and chemoprevention Konstantin H. Dragnev1,2 , W. Jeffrey Petty2 , Yan Ma6 , Qing Feng6 , Ian Pitha-Rowe6 , Bernard Beaulieu2 , Vincent Memoli1,3 , Ian Williams1 , William Black1,4 , William C. Nugent1,5 , Lionel Lewis1,2,6 , Kenneth K. Iwata7 , David Sekula6 , Sarah Freemantle6 , James Rigas1,2 , Ethan Dmitrovsky1,2,6 1 Norris Cotton Cancer Center, Dartmouth Medical School, Hanover, NH, USA; 2 Department of Medicine, 3 Department of Pathology, 4 Department of Radiology, 5 Department of Surgery, 6 Department of Pharmacology and Toxicology, DartmouthHitchcock Medical Center, Lebanon, NH, USA; 7 OSI Pharmaceuticals, Melville, NY 11553-3649, USA Multi-step carcinogenesis is an accepted tenet in molecular oncology. However, what genetic alterations are required for the maintenance and progression of a preneoplastic or neoplastic lesion in the lung and at other organ sites are not yet known. What needs to be determined is whether alterations present in preneoplastic and neoplastic lesions are also candidate therapeutic or chemopreventive targets. We set out to study this by examining treatment of immortalized human bronchial epithelial (HBE) cells with tobacco-derived carcinogens. These treatments independently caused malignant transformation of these HBE cells and enhanced expression of cyclin D1 and the epidermal growth factor receptor (EGFR). Notably, treatment with the chemopreventive agent, all-trans-retinoic acid (RA), prevented this transformation and suppressed expression of each of these species, but through different mechanisms. RA engaged transcriptional repression of EGFR and a post-transcriptional repression of cyclin D1 that involved a proteasome-dependent degradation mechanism. Similar effects were noted following treatment with a non-classical retinoid that activated the retinoid X receptor (RXR) or rexinoid pathway. This would by-pass repression of the retinoic acid receptor ␤ (RAR␤) that is frequently detected in lung carcinogenesis. To determine whether cyclin D1 or EGFR were therapeutic targets in lung carcinogenesis, a proof of principle trial was conducted using the EGFR tyrosine kinase inhibitor (EGFR-TKI) erlotinib (tarceva). This trial was conducted in aerodigestive tract

Abstracts cancers. Plasma and tissue pharmacokinetics were performed. Changes in EGFR, phospho-EGFR, cyclin D1, and Ki-67 expression were studied in pretreatment as compared to post-treatment biopsies. This erlotinib trial confirmed that in responding cases (determined by induction of necrosis in post-treatment biopsies), a marked repression of cyclin D1 and Ki-67 expression was observed. In contrast, these changes were undetected in nonresponding cases. Non-responding cases also had much lower erlotinib concentrations in tumor tissues than did responding cases, indicating a relationship existed between these drug concentrations and anti-tumor responses. Similar plasma erlotinib concentrations were measured in responding and non-responding cases. Analyses are underway for an RXR agonist (bexarotene) proof of principle trial conducted in resected non-small cell lung cancer (NSCLC) cases. Results of a phase I/II trial in refractory aerodigestive tract cancers of erlotinib combined with bexarotene will be presented. Gene profiling experiments were also conducted using carcinogen-exposed HBE cells. Novel molecular therapeutic or chemopreventive targets were uncovered that would affect cyclin D1 stability, as will be discussed. Thus, findings that will be presented highlight cyclin D1 and EGFR as molecular targets in lung carcinogenesis. Whether these targets play a pharmacological role in lung cancer therapy and chemoprevention is the subject of current work. Dose-escalation utilizing 3D planning in the combined modality approach to unresectable stage III non-small cell lung cancer Mark A. Socinski, M.D., Julian G. Rosenman, Ph.D., M.D. Multidisciplinary Thoracic Oncology Program, Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC, USA Combined modality therapy has evolved as the standard of care in the treatment of stage IIIA/B unresectable non-small cell lung cancer (NSCLC) [1]. Both sequential as well as concurrent chemoradiotherapy strategies have been shown to improve survival over thoracic radiation therapy (TRT) alone [2]. Recent trials comparing sequential versus concurrent approaches have suggested that concurrent chemoradiotherapy modestly improves survival outcomes at the risk of increased toxicity mainly in the form of myelosuppression and esophagitis [3—6]. This increase risk of toxicity occurs mainly in the acute phase with late toxicity rates appearing to be equivalent between the two approaches [4]. Although most physicians

Abstracts have been convinced of the benefits of concurrent therapy, the nature of the concurrent aspect of therapy varies between full-dose versus low-dose chemotherapy. The former may have both locoregional as well as systemic effects while the latter is essentially a radio-enhancing strategy and is probably inadequate by itself based on a recent phase III trial presented by the Cancer and Leukemia Group B (CALGB) [7]. Loco-regional control remains a formidable challenge in the unresectable stage III NSCLC patient. In a landmark analysis of recurrence patterns, Arriagada and colleagues have shown that local control rates at standard doses of TRT are suboptimal [8]. At a dose of 65 Gy, local control was achieved in less than 15% of patients treated either with TRT alone or sequential chemoradiotherapy. Strategies to improve local control include concurrent use of chemotherapy as radiosensitization or various radiotherapy strategies such as enhanced tumor targeting (three-dimensional (3D) conformal planning), dose escalation, fractionation and or acceleration. The current standard TRT dose of 60 Gy was defined by Radiation Therapy Oncology Group (RTOG) 73-01(9). Subsequent attempts at TRT dose-escalation have been relatively limited secondary to concerns about prohibitive toxicity. The advent of CT-based planning systems allowing three-dimensional approaches fostered several dose-escalation efforts in NSCLC. This technology allows for enhanced tumor targeting as well as the ability to limit excessive dose to normal tissues thereby reducing the risk for severe toxicities. Several dose-escalation trials have been performed at the Lineberger Comprehensive Cancer Center (LCCC), University of North Carolina using 3D planning in the combined modality therapy of stage III NSCLC. Our approach in general has been to use limited elective nodal irradiation in all of our studies. The first trial (LCCC 9603) used both induction as well as concurrent carboplatin/paclitaxel (C/P) along with 3D TRT planning [10]. The concurrent C/P was delivered on a low-dose weekly schedule. This was a phase I/II trial escalating the dose of TRT from 60 Gy to 74 Gy. Sixty-two stage IIIA/B patients were entered all but one of which was performance status (PS) 0—1. Induction C/P was well tolerated with a 40% response rate after two cycles. The dose of TRT was escalated from 60 Gy to 74 Gy without dose-limiting toxicity. The 74 Gy dose level was expanded in a phase II fashion (32 patients) to better define the tolerability of this higher dose level. Overall, the rate of grade 3—4 esophagitis was only 8% and no high grade pnemonitis was observed. The overall response rate was 50% and the median survival was 26 months with 1-, 3-, and 5-year sur-

S27 vival rates of 71%, 39% and 26%, respectively [11]. In an analysis of the technical aspects of this trial [12], the length of esophagus contained in the high dose field correlated with severe esophagitis with patients having >13.5 cm of esophagus in the high dose field (>60 Gy) at risk (P = 0.024). It should also be noted that the volume of normal lung receiving >20 Gy (V20 ) was <35% in nearly all these patients. Both loco-regional and systemic failures occurred with nearly equal frequency justifying future strategies which addressed both of these issues. Given the relatively favorable survival and tolerability experience seen on LCCC 9603, we embarked on a subsequent trial (LCCC 2001) which continued our dose-escalation escalation efforts [13]. We employed a more aggressive induction regimen consisting of carboplatin, irinotecan and paclitaxel (CIP) [14] and continued the dose-escalation of TRT using 3D planning from 78 Gy to 90 Gy again with concurrent low-dose weekly C/P. Twenty-nine patients with stage IIIA/B NSCLC and PS 0—1 were entered. All patients had V20 values <40% and favorable dose-length parameters regarding the esophagus (median length of esophagus in the high dose field was 6.5 cm). The induction CIP regimen was well tolerated and active with a 40% response rate. The dose of TRT was escalated successfully from 78 Gy to 90 Gy without dose-limiting toxicity. One case of grade 3 radiation pnemonitis was seen at the 90 Gy dose level in a patient whose V20 approached 40%. Unfortunately, two cases of fatal late pulmonary hemorrhage occurred (one at 82 Gy and one at 90 Gy) and one patient developed a bronchoesophageal fistula (at 78 Gy). The overall response rate was 60% and the median survival is 24 months. A third trial (LCCC 9732) was performed in collaboration with investigators at Duke University and Wake Forest University (3D Carolina Consortium) [15]. This trial used a novel “ping-pong” design evaluating two induction regimens (C/P and carboplatin/vinorelbine) followed by hyperfractionated TRT delivered utilizing a concurrent boost technique. No concurrent chemotherapy was used. Following two cycles of induction therapy, patients received dose-escalating 3D planned TRT with 1.6 Gy and 1.25 Gy bid delivered to the gross and clinical target volumes, respectively. The dose of TRT was escalated from 73.6 Gy to 86.4 Gy in 6.4 Gy increments. Forty-four patients with stage IIIA/B were entered. The maximum tolerated dose using this technique was 80 Gy with dose limiting toxicities consisting mainly of pulmonary and esophageal events. The overall median survival was 18 months. Several other investigators have explored doseescalation or dose-escalated approaches using 3D

S28 planning [16—21]. In general, these trials have included varied patient populations (stage I—III) and inconsistent chemotherapy approaches. The TRT approaches have also varied in terms of total dose as well as the use of elective nodal irradiation. Having stated these limitations, the current literature would support the contention that doses in excess of 70 Gy are both feasible and safe in the general population of unresectable stage III NSCLC patients. This assumption is based on the general adherence to the planning strategies outlined in the individual reports. The more important question is whether or not a higher dose of TRT would impact overall survival. This issue could only be addressed in a randomized phase III trial.

References [1] Pfister DG, Johnson DH, Azzoli CG, et al. American Society of clinical oncology treatment of unresectable nonsmall cell lung cancer guideline: update 2003. J Clin Oncol 2004;22(2):330—53. [2] Hensing T, Halle J, Socinski MA. Chemoradiotherapy for stage IIIA, B non-small cell lung cancer. In: Detterbeck FC, Rivera MP, Socinski MA, Rosenman JG, editors. Diagnosis and treatment of lung cancer: an evidence-based guide for the practicing clinician. Philadelphia, PA: WB Saunders; 2001. [3] Furuse K, for the West Japan Lung Cancer Group. Phase III study of concurrent versus sequential thoracic radiotherapy in combination with mitomycin, vindesine, and cisplatin in unresectable stage III non-small cell lung cancer. J Clin Oncol 1999;17:2692—9. [4] Curran D, Scott C, Langer C, et al. Long-term benefit is observed in a phase III comparison of sequential versus concurrent chemo-radiation for patients with unresected stage III NSCLC: RTOG 9410. Proc Am Soc Clin Oncol 2003;22:621. [5] Pierre F, Maurice P, Gilles R, et al. A randomized phase III trial of sequential chemo-radiotherapy versus concurrent chemo-radiotherapy in locally advanced non-small cell lung cancer (NSCLC) (GLOT-GFPC NPC 95-01 study). Proc Am Soc Clin Oncol 2001;20:312 (abstract 1246). [6] Zatloukal P, Petruzelka L, Zemanova M, et al. Concurrent versus sequential radiochemotherapy with vinorelbine plus cisplatin (V—P) in locally advanced non-small cell lung cancer: a randomized phase II study. Proc Am Soc Clin Oncol 2002;21:290a. [7] Vokes EE, Herndon JE, Kelley MJ, et al. Induction chemotherapy followed by concomitant chemoradiotherapy (CT/XRT) versus CT/XRT alone for regionally advanced unresectable non-small cell lung cancer (NSCLC): initial analysis of a randomized trial phase III trial. Proc Am Soc Clin Oncol 2004;23, 616 (abstract 7005). [8] Arriagada R, Le Chevalier T, Quoix E. ASTRO Plenary: effect of chemotherapy on locally advanced non-small cell lung carcinoma: a randomized study of 353 patients. Int J Radiat Oncol Biol Phys 1994;20:537—44. [9] Perez CA, Stanley K, Grundy G, et al. Impact of irradiation technique and tumor extent in tumor control and survival of patients with unresectable non-oat cell carcinoma of the lung: report by the Radiation Therapy Oncology Group. Cancer 1982;50(6):1091—9.

Abstracts [10] Socinski MA, Rosenman J, Halle J, et al. Dose-escalating conformal thoracic radiation therapy with induction and concurrent carboplatin/paclitaxel in unresectable stage IIIA/B non-small cell lung carcinoma: a modified phase I/II trial. Cancer 2001;92:1213—23. [11] Socinski MA, Halle JS, Morris DE, et al. Long-term results of aggressive combined modality therapy employing induction and concurrent carboplatin/paclitaxel with doseescalated thoracic conformal radiation therapy. Lung Cancer 2003;41(S2):S239. [12] Rosenman JG, Halle JS, Socinski MA, et al. High-dose conformal radiotherapy for treatment of stage IIIA/IIIB nonsmall-cell lung cancer: technical issues and results of a phase I/II trial. Int J Radiat Oncol Biol Phys 2002;54(2):348— 56. [13] Socinski MA, Rosenman JG, Halle J, et al. Induction and concurrent chemotherapy with dose-escalated thoracic conformal radiation therapy (TCRT) in unresectable stage iiia/b non-small cell lung cancer. Lung Cancer 2003;41(S2):S90. [14] Socinski MA, Sandler A, Miller L, et al. Phase I trial of the combination of irinotecan (CPT-11), paclitaxel, and carboplatin in patients with advanced non-small cell lung cancer. J Clin Oncol 2001;19:1078—87. [15] Marks LB, Garst J, Socinski MA, et al. Carboplatin/paclitaxel (c/t) or carboplatin/vinorelbine (c/n) followed by accelerated hyperfractionated conformal radiation therapy (R/T): report of a prospective phase I dose escalation trial from the carolina conformal therapy consortium. Proc Am Soc Clin Oncol 2003;22:637. [16] Maguire PD, Marks LB, Sibley GS, et al. 73.6 Gy and beyond: hyperfractionated, accelerated radiotherapy for non-smallcell lung cancer. J Clin Oncol 2001;19(3):705—11. [17] Sim S, Rosenzweig KE, Schindelheim R, et al. Induction chemotherapy plus three-dimensional conformal radiation therapy in the definitive treatment of locally advanced non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2001;51(3):660—5. [18] Hayman JA, Martel MK, Ten Haken RK, et al. Dose escalation in non-small-cell lung cancer using three-dimensional conformal radiation therapy: update of a phase I trial. J Clin Oncol 2001;19(1):127—36. [19] Bradley JD, Graham MV, Winter KW, et al. Acute and late toxicity results of RTOG 9311: a dose escalation study using 3D conformal radiation therapy in patients with inoperable non-small cell lung cancer. Int J Radiat Oncol Biol Phys 2003;57(Suppl 2):S137—8. [20] Wu KL, Jiang GL, Liao Y, et al. Three-dimensional conformal radiation therapy for non-small-cell lung cancer: a phase I/II dose escalation clinical trial. Int J Radiat Oncol Biol Phys 2003;57(5):1336—44. [21] Belderbos JS, De Jaeger K, Heemsbergen WD, et al. First results of a phase I/II dose escalation trial in non-small cell lung cancer using three-dimensional conformal radiotherapy. Radiother Oncol 2003;66(2):119—26.

Emerging role of targeted therapy in the management of locally advanced (LA) non-small cell lung cancer (NSCLC) Corey J. Langer, M.D., FACP Thoracic Oncology, Fox Chase Cancer Center, Philadelphia, PA, USA Combined modality therapy has evolved as the standard of care for fit patients with LA-NSCLC. At least

Abstracts seven clinical trials have shown a survival advantage for sequential or concurrent chemoradiation (RT) over RT alone [1—8]; and three out of four recent clinical trials have demonstrated therapeutic superiority for concurrent chemoRT versus sequential therapy [8—12]. Unfortunately, we have witnessed a plateau in median and long-term survival rates. In randomized trials, median survival seldom exceeds 17—18 months, and 5-year survival rates in the two largest trials are only 16% [9—11]. The SWOG 9504 trial has suggested that consolidation with a putative “non-cross resistant regimen” can improve median survival rate beyond 2 years and increase 3-year survival rates to nearly 40%; but these observations are based on a phase II trial with fewer than 100 patients enrolled [13]. There are no randomized data to confirm the benefits of consolidative cytotoxics. Hence, the careful, deliberate evaluation of newer targeted agents in the setting of LA-NSCLC is rational and offers the prospect of improved survival without excess toxicity. The most actively investigated class(es) of agents include inhibitors of epidermal growth factor receptor (EGFr), angiogenesis, and cyclo-oxygenase 2 (Cox-2). EGFr expression has been associated with impaired survival in NSCLC and increase risk of metastatic spread. Agents inhibiting EGFr may potentiate RT by inducing G1 arrest, promoting apoptosis, inhibiting DNA-damage repair and tumor neoangiogenesis, and stopping cellular proliferation [14]. In A549, a NSCLC cell line, C225 (erbitux) has demonstrated synergy in combination with RT, cisplatin and vinorelbine, and additivity with paclitaxel at higher doses [15]. The treatment of cultured human SCCHN cells with C225 has been shown to heighten radio-response in vitro; this has been attributed to RT-induced apoptosis. In a landmark phase III trial evaluating the role of C225 in locally advanced squamous cell carcincoma of the head and neck, the addition of C225 to standard RT led to a significant survival benefit compared to RT alone [16]. Efforts to assess the role of EGFr inhibition in LA-NSCLC are fledgling thus far. Blumenshein et al. of the RTOG is assessing the role of C225 grafted onto a standard regimen of concurrent chemoradiation with weekly radiosensitizing doses of paclitaxel and carboplatin followed by concurrent C225, and two additional cycles of chemotherapy. To date (6/04), four patients have been accrued; there are no reports of untoward toxicity. The CALGB is mounting a phase II trial assessing the role of gefitinib in LA-NSCLC. All patients receive induction therapy with paclitaxel and carboplatin in combination with gefitinib. Fit patients (ECOG PS 0—1 and <5 wt.% loss)

S29 go on to receive concurrent chemoradiation with daily gefitinib, while more vulnerable patients (PS 2 or >5 wt.% loss) receive RT and gefitinib alone. To date, only one of six patients receiving concurrent chemoradiation and gefitinib has experienced dose-limiting toxicity; there has been no untoward toxicity in the poor risk cohort [17]. Rischin and colleagues from Australia [18] have also demonstrated the feasibility of combining gefitinib with RT and escalating doses of carboplatin and paclitaxel. The University of Colorado is assessing the role of escalating doses of 3D conformal RT in combination with paclitaxael and carboplatin and daily gefitinib upfront followed by additional two cycles of chemotherapy and gefitinib continued for up to 2 years. Investigators at the University of Chicago are evaluating the role of erlotinib, another small molecule EGFr TKI, in combination with RT and either concurrent etoposide/cisplatin or paclitaxel/carboplatin. Phase II trials of standard chemotherapy with bevacizumab have suggested therapeutic promise, though concerns exist regarding sudden, occasionally fatal tumor hemorrhage. Trial combining bevacizumab with chemoradiation are being planned. Thalidomide, a putative angiogenesis inhibitor, is being evaluated by the ECOG in the context of LA-NSCLC. E3598 randomizes patients to induction chemotherapy followed by standard RT either with or without thalidomide. This trial, which has accrued over 350 patients to date, was recently amended to allow concurrent chemoradiation. A consortium led by MDACC is running a similar trial with Aeterna, the shark cartilage compound, which has been purported to act by a similar mechanism. Finally, celecoxib is undergoing assessment in the setting of LA-NSCLC. Cox-2 inhibition can abrogate or reverse up-regulated Cox-2 expression after RT, leading to enhanced apoptosis [19,20]. RTOG is spearheading a phase II trial of concurrent RT and celecoxib in poor risk patients, and the VCCAN network is testing celecoxib in combination with concurrent chemo-radiation with paclitaxel and carboplatin. Over 13 patients have been accrued to the former effort with no reports of untoward toxicity; 9 have been enrolled in the latter effort with 2 patients experiencing grade 3 pneumonitis and 3 experiencing grade 3 or 4 esophagitis [21]. Two grade 5 pulmonary events have occurred, but it is not clear if they are related in any way to celecoxib. Although long-term data are not yet available for these efforts, the simple realignment of cytotoxics is not likely to improve long-term outcome in the therapy of LA-NSCLC; hence, targeted agents, particularly those focussing on Cox-2, EGFr, and angiogenesis offer hope, if not promise. Of course, phase

S30 III trials will be needed before we can determine if these agents enhance long-term survival.

References [1] Dillman RO, Herndon J, Seagren SL, Eaton WL, Green MR. Improved survival in stage III non-small cell lung cancer: seven-year follow-up of Cancer and Leukemia Group B (CALGB) 8433 trial. J Natl Cancer Inst 1996;88: 1210—5. [2] Sause WT, Scott C, Taylor S, et al. Radiation Therapy Oncology Group (RTOG) 88-08 and Eastern Cooperative Oncology Group (ECOG) 4588: preliminary results of a phase III trial in regionally advanced, unresectable non-small cell lung cancer. J Natl Cancer Inst 1995;87:198—205. [3] LeChevalier T, Arriagada R, Lacombe-Terrier MJ, Laplanche A. Significant effect of adjuvant chemotherapy on survival in locally advanced non-small cell lung carcinoma. J Natl Clin Cancer Inst 1992;84:58. [4] Sause WT, Kolesar P, Taylor SG, Johnson D, Livingston R, Komaki R, et al. Five-year results: phase III trial of regionally advanced unresectable non-small cell lung cancer, RTOG 8808, ECOG 4588; SWOG 8992. Proc ASCO 1998;17: A1743. [5] Cullen MH, Billingham LJ, Woodroffe CM, et al. Mitomycin, ifosfamide, and cisplatin in unresectable non-small cell lung cancer: effects on survival and quality of life. J Clin Oncol 1999;17:3188—94. [6] Schaake-Koning C, van den Bogaert W, Dalesio O et al. Effects of concomitant cisplatin and radiotherapy on inoperable non-small cell lung cancer. N Engl J Med 1992;326:524—30. [7] Jeremic B, Shibamoto Y, Acimovic L, Djuric L. Randomized trial of hyperfractionated radiation therapy with or without concurrent chemotherapy for stage III non-small cell lung cancer. J Clin Oncol 1995;13:452—8.

Abstracts [8] Jeremic B, Shibamoto Y, Acimovic L, Milisarljenic S. Hyperfractionated radiation therapy with or without concurrent low-dose daily carboplatin-etoposide for stage III non-small cell lung cancer. J Clin Oncol 1996;14:1065— 70. [9] Furuse K, Fukuoka F, Kawahara M, Nishikawa H, Takada Y, Kudoh S, et al. Phase III study of concurrent versus sequential thoracic radiotherapy in combination with mitomycin, vindesine, and cisplatin in unresectable non-small cell lung cancer. J Clin Oncol 1999;17:2692—9. [10] Curran W, Scott C, Langer C, et al. Phase III comparison of sequential versus concurrent chemoradiation for patients with unresectable, stage III NSCLC: initial report of RTOG 9410. Proc Am Soc Clin Oncol 2000 (abstract A-1891). [11] Curran WJ, Scott CB, Langer CJ, Komaki R, Lee JS, Hauser S, et al. Long-term benefit is observed in a phase III comparison of sequential versus concurrent chemoradiation for patients with unresected stage III NSCLC. Proc Am Soc Clinic Oncol 2003;22(A-2499):621. [12] Zemanova M, Petruzelka L, Zemanova M, et al. Concurrent versus sequential radiochemotherapy with vinorelbine plus cisplatin (V—P) in locally advanced NSCLC: a randomized phase II study. Proc Am Soc Clin Oncol. 2002;21(A1159):290a. [13] Gandara DR. Proc Am Soc Clin Oncol 2000;19(A-1916). [14] Balaban. Biochem Biophys 1996;1314:147. [15] Milas L, Mason K, Hunter N, Petersen S, Yamakawa M, Ang K, et al. In vivo enhancement of tumor radio-response by C225 anti-epidermal growth factor receptor antibody. Clin Cancer Res 2000;6:701—8. [16] Bonner, et al. Proc Am Soc Clin Oncol 2004 (abstract). [17] Ready, et al. Proc Am Soc Clin Oncol 2004 (abstract 7078). [18] Rischin, et al. Proc Am Soc Clin Oncol 2004 (abstrcat 7077). [19] Steinhauer IJROBP 2000; Gaffney IJROBP 2001. [20] Kishi, et al. IJROBP 2001. [21] Carbone, et al. Proc Am Soc Clin Oncol 2002;21:318a (abstract 1270).