Targeting EGFR in Non–Small-Cell Lung Cancer
http://www.100md.com
《新英格兰医药杂志》
The development of small-molecule inhibitors of the epidermal growth factor receptor (EGFR), such as erlotinib and gefitinib, for the treatment of advanced non–small-cell lung cancer illustrates a pattern of initial overly enthusiastic interest, subsequent critical disappointment, and eventual renewed appreciation of the clinical and mechanistic complexities surrounding the introduction of a new therapy for cancer. Such a pattern has been regrettably common in the history of oncologic therapeutics.
The initial safety and efficacy studies of erlotinib and gefitinib evoked a remarkable degree of optimism that was based on prolonged remissions and, in some cases, dramatic improvements in the quality of life in small numbers of patients whose condition was no longer responding to standard chemotherapy.1,2 Furthermore, clinical benefit was achieved by treatments that appeared to control, with acceptable levels of toxic effects, advanced non–small-cell lung cancer through a specific molecular mechanism. However, when in randomized clinical trials EGFR inhibitors were combined with cytotoxic chemotherapy, no advantage was demonstrated over standard chemotherapy alone — a result that provoked substantial consternation regarding the potential value of this new class of agents in the treatment of non–small-cell lung cancer and doubts about the clarity of our knowledge of their presumed mechanism of action.3,4
In 2004, the public presentation of clinical trial BR.21 conducted by the National Cancer Institute of Canada (NCIC) showing a two-month survival advantage with erlotinib, as is reported in full in this issue of the Journal,5,6 had a positive reception that was particularly warranted because many of the participants had previously received two chemotherapy regimens. As a group, such patients are unlikely to benefit from further cytotoxic approaches. At almost the same moment, interest in EGFR inhibitors was heightened by reports linking mutations in exons 18 through 21 of the EGFR gene to the responsiveness of non–small-cell lung cancer to gefitinib.7,8 Taken together, these findings rekindled substantial enthusiasm for the use of EGFR inhibitors in the treatment of non–small-cell lung cancer. The result has been a flurry of investigative activity to determine the role of these mutations, and potentially that of other alterations in EGFR signaling pathways, in selecting patients who might benefit from EGFR-inhibitor therapy.9
The two reports from the NCIC in this issue of the Journal, by Shepherd and colleagues5 on the clinical results and by Tsao and colleagues6 on the molecular results of a randomized trial comparing erlotinib with placebo in non–small-cell lung cancer, provide data that have a substantial bearing on the treatment of patients with advanced lung cancer. From these studies, critical lessons can also be derived on how anticancer agents that have been selected to affect a specific molecular pathway should be developed.
The NCIC trial is the first randomized study of a tyrosine kinase inhibitor to demonstrate substantial benefit in terms of both relief of symptoms (cough, dyspnea, and pain) and median and one-year survival in non–small-cell lung cancer. This outcome was achieved with mild toxic effects with the use of an oral antineoplastic agent (erlotinib) and led to its approval for use in non–small-cell lung cancer by the Food and Drug Administration in November 2004. Patient characteristics that in phase 2 studies of non–small-cell lung cancer had been associated with responsiveness to EGFR inhibitors — histologic features of adenocarcinoma, female sex, no history of smoking, and Asian ancestry — correlated significantly with responsiveness to erlotinib in this trial.
In the accompanying study of potential molecular correlates of sensitivity to erlotinib,6 paraffin-embedded biopsy and surgical-resection samples of non–small-cell lung cancer, adequate for at least one analysis of EGFR, were available from 328 of the 731 patients enrolled. Response to erlotinib was significantly associated with EGFR positivity on immunohistochemical analysis and increased numbers of copies of the EGFR gene. Although twice as many patients with EGFR mutations had a response to treatment, as compared with patients whose tumors lacked mutations, this difference was not statistically significant. In the univariate analysis, a survival advantage was associated with increased numbers of gene copies or EGFR expression in patients treated with erlotinib; however, this association was not found in the multivariate analyses.
The importance of these results is heightened by the fact that molecular correlations were performed in the context of a randomized, placebo-controlled study. The observation that the clinical benefit from treatment with erlotinib found in all subgroups that underwent the EGFR analysis was similar to that in the entire study population suggests that the distribution of tumor samples within the subgroups was representative of all patients entered into the trial. In support of the NCIC findings, the role of an increase in the numbers of copies of the EGFR gene as a potential marker of sensitivity to gefitinib in non–small-cell lung cancer has recently been demonstrated by others.10
These results, however, raise critical new questions with regard to the mechanism of action of erlotinib in non–small-cell lung cancer and the methodologic requirements for molecular pharmacodynamic studies in the future. Although increases in the number of gene copies, protein expression, and EGFR mutations were associated with a response, objective responses to erlotinib were also found in the absence of these features. This finding suggests that erlotinib may affect, or be dependent for its activity on, additional signaling pathways that were not assessed in the NCIC study (such as Akt phosphorylation or erbB-3 status) and that are essential for tumor-cell proliferation.10,11 Thus, further investigation to determine the mechanisms by which erlotinib alters the full range of EGFR-related signaling pathways is essential and is likely to lead to the discovery of a panel of molecular markers that will eventually be used to predict the therapeutic activity of this class of drugs in non–small-cell lung cancer.
Perhaps of greatest importance is the fact that samples of tumor tissues from fewer than half the patients who participated in the NCIC trial were usable for EGFR analysis. In particular, the small number of samples that could be examined for EGFR mutations or number of gene copies clearly reduced the strength of the conclusions it was possible to draw from this study. Furthermore, the tissues that were available were taken from either initial surgical specimens or initial or subsequent biopsies performed before the initiation of therapy. In the absence of a standardized approach to ensuring tumor acquisition at entry into the study, the effect of tumor size and time from diagnosis to first treatment on measures of EGFR could not be controlled for in the trial.
With the development of trastuzumab therapy for breast cancer as a model,12 it is now clear that predictive molecular assays must be devised before the initiation of clinical trials for new targeted anticancer agents if the specificity and usefulness of these drugs are to be meaningfully evaluated in the population of patients most likely to benefit from the treatment. Furthermore, the designs of modern phase 2 and phase 3 clinical trials must incorporate specific molecular assays to maximize the likelihood of definitive clinical results. This work will require substantially more effort to develop standardized assay procedures for assessing and predicting the effects of whole classes of molecularly targeted agents and highly standardized methods for the prospective collection of human tumor tissue.
The results of the NCIC study of erlotinib for non–small-cell lung cancer support the promise that individualized oncologic therapies can be developed and applied on the basis of the specific molecular characteristics of a patient's malignant disease. Only the most assiduous attention to the development of drug-specific molecular pharmacodynamic tools, and the implementation of meticulous tumor-acquisition techniques, will allow that promise to be realized.
Source Information
From the Division of Cancer Treatment and Diagnosis, National Cancer Institute, Bethesda, Md.
References
Kris MG, Natale RB, Herbst RS, et al. Efficacy of gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: a randomized trial. JAMA 2003;290:2149-2158.
Perez-Soler R, Chachoua A, Hammond LA, et al. Determinants of tumor response and survival with erlotinib in patients with non-small-cell lung cancer. J Clin Oncol 2004;22:3238-3247.
Herbst RS, Giaccone G, Schiller JH, et al. Gefitinib in combination with paclitaxel and carboplatin in advanced non-small-cell lung cancer: a phase III trial -- INTACT 2. J Clin Oncol 2004;22:785-794.(James H. Doroshow, M.D.)
The initial safety and efficacy studies of erlotinib and gefitinib evoked a remarkable degree of optimism that was based on prolonged remissions and, in some cases, dramatic improvements in the quality of life in small numbers of patients whose condition was no longer responding to standard chemotherapy.1,2 Furthermore, clinical benefit was achieved by treatments that appeared to control, with acceptable levels of toxic effects, advanced non–small-cell lung cancer through a specific molecular mechanism. However, when in randomized clinical trials EGFR inhibitors were combined with cytotoxic chemotherapy, no advantage was demonstrated over standard chemotherapy alone — a result that provoked substantial consternation regarding the potential value of this new class of agents in the treatment of non–small-cell lung cancer and doubts about the clarity of our knowledge of their presumed mechanism of action.3,4
In 2004, the public presentation of clinical trial BR.21 conducted by the National Cancer Institute of Canada (NCIC) showing a two-month survival advantage with erlotinib, as is reported in full in this issue of the Journal,5,6 had a positive reception that was particularly warranted because many of the participants had previously received two chemotherapy regimens. As a group, such patients are unlikely to benefit from further cytotoxic approaches. At almost the same moment, interest in EGFR inhibitors was heightened by reports linking mutations in exons 18 through 21 of the EGFR gene to the responsiveness of non–small-cell lung cancer to gefitinib.7,8 Taken together, these findings rekindled substantial enthusiasm for the use of EGFR inhibitors in the treatment of non–small-cell lung cancer. The result has been a flurry of investigative activity to determine the role of these mutations, and potentially that of other alterations in EGFR signaling pathways, in selecting patients who might benefit from EGFR-inhibitor therapy.9
The two reports from the NCIC in this issue of the Journal, by Shepherd and colleagues5 on the clinical results and by Tsao and colleagues6 on the molecular results of a randomized trial comparing erlotinib with placebo in non–small-cell lung cancer, provide data that have a substantial bearing on the treatment of patients with advanced lung cancer. From these studies, critical lessons can also be derived on how anticancer agents that have been selected to affect a specific molecular pathway should be developed.
The NCIC trial is the first randomized study of a tyrosine kinase inhibitor to demonstrate substantial benefit in terms of both relief of symptoms (cough, dyspnea, and pain) and median and one-year survival in non–small-cell lung cancer. This outcome was achieved with mild toxic effects with the use of an oral antineoplastic agent (erlotinib) and led to its approval for use in non–small-cell lung cancer by the Food and Drug Administration in November 2004. Patient characteristics that in phase 2 studies of non–small-cell lung cancer had been associated with responsiveness to EGFR inhibitors — histologic features of adenocarcinoma, female sex, no history of smoking, and Asian ancestry — correlated significantly with responsiveness to erlotinib in this trial.
In the accompanying study of potential molecular correlates of sensitivity to erlotinib,6 paraffin-embedded biopsy and surgical-resection samples of non–small-cell lung cancer, adequate for at least one analysis of EGFR, were available from 328 of the 731 patients enrolled. Response to erlotinib was significantly associated with EGFR positivity on immunohistochemical analysis and increased numbers of copies of the EGFR gene. Although twice as many patients with EGFR mutations had a response to treatment, as compared with patients whose tumors lacked mutations, this difference was not statistically significant. In the univariate analysis, a survival advantage was associated with increased numbers of gene copies or EGFR expression in patients treated with erlotinib; however, this association was not found in the multivariate analyses.
The importance of these results is heightened by the fact that molecular correlations were performed in the context of a randomized, placebo-controlled study. The observation that the clinical benefit from treatment with erlotinib found in all subgroups that underwent the EGFR analysis was similar to that in the entire study population suggests that the distribution of tumor samples within the subgroups was representative of all patients entered into the trial. In support of the NCIC findings, the role of an increase in the numbers of copies of the EGFR gene as a potential marker of sensitivity to gefitinib in non–small-cell lung cancer has recently been demonstrated by others.10
These results, however, raise critical new questions with regard to the mechanism of action of erlotinib in non–small-cell lung cancer and the methodologic requirements for molecular pharmacodynamic studies in the future. Although increases in the number of gene copies, protein expression, and EGFR mutations were associated with a response, objective responses to erlotinib were also found in the absence of these features. This finding suggests that erlotinib may affect, or be dependent for its activity on, additional signaling pathways that were not assessed in the NCIC study (such as Akt phosphorylation or erbB-3 status) and that are essential for tumor-cell proliferation.10,11 Thus, further investigation to determine the mechanisms by which erlotinib alters the full range of EGFR-related signaling pathways is essential and is likely to lead to the discovery of a panel of molecular markers that will eventually be used to predict the therapeutic activity of this class of drugs in non–small-cell lung cancer.
Perhaps of greatest importance is the fact that samples of tumor tissues from fewer than half the patients who participated in the NCIC trial were usable for EGFR analysis. In particular, the small number of samples that could be examined for EGFR mutations or number of gene copies clearly reduced the strength of the conclusions it was possible to draw from this study. Furthermore, the tissues that were available were taken from either initial surgical specimens or initial or subsequent biopsies performed before the initiation of therapy. In the absence of a standardized approach to ensuring tumor acquisition at entry into the study, the effect of tumor size and time from diagnosis to first treatment on measures of EGFR could not be controlled for in the trial.
With the development of trastuzumab therapy for breast cancer as a model,12 it is now clear that predictive molecular assays must be devised before the initiation of clinical trials for new targeted anticancer agents if the specificity and usefulness of these drugs are to be meaningfully evaluated in the population of patients most likely to benefit from the treatment. Furthermore, the designs of modern phase 2 and phase 3 clinical trials must incorporate specific molecular assays to maximize the likelihood of definitive clinical results. This work will require substantially more effort to develop standardized assay procedures for assessing and predicting the effects of whole classes of molecularly targeted agents and highly standardized methods for the prospective collection of human tumor tissue.
The results of the NCIC study of erlotinib for non–small-cell lung cancer support the promise that individualized oncologic therapies can be developed and applied on the basis of the specific molecular characteristics of a patient's malignant disease. Only the most assiduous attention to the development of drug-specific molecular pharmacodynamic tools, and the implementation of meticulous tumor-acquisition techniques, will allow that promise to be realized.
Source Information
From the Division of Cancer Treatment and Diagnosis, National Cancer Institute, Bethesda, Md.
References
Kris MG, Natale RB, Herbst RS, et al. Efficacy of gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: a randomized trial. JAMA 2003;290:2149-2158.
Perez-Soler R, Chachoua A, Hammond LA, et al. Determinants of tumor response and survival with erlotinib in patients with non-small-cell lung cancer. J Clin Oncol 2004;22:3238-3247.
Herbst RS, Giaccone G, Schiller JH, et al. Gefitinib in combination with paclitaxel and carboplatin in advanced non-small-cell lung cancer: a phase III trial -- INTACT 2. J Clin Oncol 2004;22:785-794.(James H. Doroshow, M.D.)