DNA Repair by ERCC1 in Non–Small-Cell Lung Cancer
http://www.100md.com
《新英格兰医药杂志》
To the Editor: Olaussen and colleagues (Sept. 7 issue)1 report a survival benefit associated with the absence of the excision repair cross-complementation group 1 (ERCC1) protein in patients with completely resected non–small-cell lung cancer who received adjuvant cisplatin-based chemotherapy. Most of the operative specimens of ERCC1-positive tumors examined were squamous-cell carcinoma, whereas the ERCC1-negative tumors were almost evenly divided between squamous-cell carcinoma and adenocarcinoma.1 Does this skewed distribution reflect a selection bias?
Originally, the International Adjuvant Lung Cancer Trial (IALT) showed a difference in survival of 2.3 percentage points for patients with squamous-cell tumors who received cisplatin (as compared with those in the control group who did not receive cisplatin-based chemotherapy). Among patients with adenocarcinoma who received cisplatin, the survival difference was 4.9 percentage points. The absolute 5-year survival benefit was 4.1 percentage points.2 Does this difference fail to differentiate between the two histologic types? Lower expression of ERCC1 is correlated with adenocarcinoma,1 decreased survival,3 and the most benefit from adjuvant cisplatin-based therapy.1 ERCC1 expression in squamous-cell carcinoma may be more useful than the histologic type as a guide to chemotherapy.
Jane E. Wilcox, B.A.
Loyola University Chicago Stritch School of Medicine
Maywood, IL 60153
jawilcox@lumc.edu
References
Olaussen KA, Dunant A, Fouret P, et al. DNA repair by ERCC1 in non-small-cell lung cancer and cisplatin-based adjuvant chemotherapy. N Engl J Med 2006;355:983-991.
The International Adjuvant Lung Cancer Trial Collaborative Group. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N Engl J Med 2004;350:351-360.
Simon GR, Sharma S, Cantor A, Smith P, Bepler G. ERCC1 expression is a predictor of survival in resected patients with non-small cell lung cancer. Chest 2005;127:978-983.
To the Editor: The findings of Olaussen et al. underscore the importance of ERCC1 expression in the risk of relapse among patients with resected non–small-cell lung cancer and in the outcome of cisplatin-based adjuvant chemotherapy. In the IALT, 56% of the patients were treated with cisplatin plus etoposide, and the remaining 44% received cisplatin plus vinca alkaloids.1 Was the benefit of adjuvant chemotherapy observed equally in both groups? Was the improvement in survival independent of the stage of disease, since no benefit has yet been observed among patients with stage I disease?
Fabiana Cecere, M.D.
Emilio Bria, M.D.
Regina Elena National Cancer Institute
00144 Rome, Italy
Rafael Rosell, M.D.
Catalan Institute of Oncology
08916 Barcelona, Spain
rrosell@ico.scs.es
References
The International Adjuvant Lung Cancer Trial Collaborative Group. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N Engl J Med 2004;350:351-360.
To the Editor: In the report by Olaussen et al., the data were not stratified according to age, sex, smoking status, or histologic type. Will a significant survival benefit persist after adjustment for all these factors?
Tal Grenader, M.D.
Linda Shavit, M.D.
Shaare Zedek Medical Center
91031 Jerusalem, Israel
talgrenader65@hotmail.com
The authors reply: Our analyses addressed, in three steps, the question of whether the characteristics of the tumors or of the patients influenced the observed differential effect of chemotherapy on survival among patients with ERCC1-negative tumors and those with ERCC1-positive tumors. First, we verified that there was no significant interaction between the treatment effect and any of the demographic and clinical characteristics of patients included in the IALT Biology (IALT Bio) study, a subgroup of the IALT study population.1 This step was also taken in the analyses performed in the IALT. The absence of an interaction was verified in particular for such factors as histologic type, stage of disease, and the assigned treatment (P=0.64, P=0.45, and P=0.26, respectively). Second, when studying the interaction between ERCC1 status and treatment, we adjusted not only for stratification and prognostic factors but also for factors associated with ERCC1 status (histopathological type, the presence or absence of pleural invasion, and age). Third, we verified that the interaction with ERCC1 status would not disappear when any other interaction was added to the Cox model (P values for an interaction with ERCC1 status remained below 0.009 in the presence of an interaction with histologic type, stage of disease, or assigned treatment). Thus, the interaction between the ERCC1 status and treatment was robust enough to withstand the addition of any other interaction.
Wilcox points out the different histologic types of carcinoma among the patients with ERCC1-negative and ERCC1-positive tumors (P<0.001, in Table 1 of our article). As shown above, this difference does not explain the interaction between ERCC1 status and the treatment effect.
Cecere and colleagues request clarification of the role of the assigned drug and the stage of disease in the interaction between ERCC1 status and treatment. Our analysis shows that this interaction is the only one that remains significant when the model includes either of these two interactions (drug or stage of disease).
Grenader and Shavit are concerned that stratification according to age, sex, smoking status, and histologic type was not performed. Our analysis took all these factors into account, except smoking status. Data on smoking status were not collected in the trial. Smoking may affect the prognosis for patients with lung cancer and may be related to ERCC1 expression. However, since the proportion of smokers is quite high among patients with non–small-cell lung cancer (around 90%), only a very wide variation in the treatment effect between smokers and nonsmokers could explain the observed significant interaction between treatment and ERCC1 status. More important, adjustments for histologic type and sex (which are classically associated with smoking) and the corresponding interactions with treatment do not change the results.
Ariane Dunant, M.S.
Pierre Fouret, M.D., Ph.D.
Jean-Charles Soria, M.D., Ph.D.
Institut de Cancérologie Gustave Roussy
94805 Villejuif CEDEX, France
soria@igr.fr
References
The International Adjuvant Lung Cancer Trial Collaborative Group. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N Engl J Med 2004;350:351-360.
Originally, the International Adjuvant Lung Cancer Trial (IALT) showed a difference in survival of 2.3 percentage points for patients with squamous-cell tumors who received cisplatin (as compared with those in the control group who did not receive cisplatin-based chemotherapy). Among patients with adenocarcinoma who received cisplatin, the survival difference was 4.9 percentage points. The absolute 5-year survival benefit was 4.1 percentage points.2 Does this difference fail to differentiate between the two histologic types? Lower expression of ERCC1 is correlated with adenocarcinoma,1 decreased survival,3 and the most benefit from adjuvant cisplatin-based therapy.1 ERCC1 expression in squamous-cell carcinoma may be more useful than the histologic type as a guide to chemotherapy.
Jane E. Wilcox, B.A.
Loyola University Chicago Stritch School of Medicine
Maywood, IL 60153
jawilcox@lumc.edu
References
Olaussen KA, Dunant A, Fouret P, et al. DNA repair by ERCC1 in non-small-cell lung cancer and cisplatin-based adjuvant chemotherapy. N Engl J Med 2006;355:983-991.
The International Adjuvant Lung Cancer Trial Collaborative Group. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N Engl J Med 2004;350:351-360.
Simon GR, Sharma S, Cantor A, Smith P, Bepler G. ERCC1 expression is a predictor of survival in resected patients with non-small cell lung cancer. Chest 2005;127:978-983.
To the Editor: The findings of Olaussen et al. underscore the importance of ERCC1 expression in the risk of relapse among patients with resected non–small-cell lung cancer and in the outcome of cisplatin-based adjuvant chemotherapy. In the IALT, 56% of the patients were treated with cisplatin plus etoposide, and the remaining 44% received cisplatin plus vinca alkaloids.1 Was the benefit of adjuvant chemotherapy observed equally in both groups? Was the improvement in survival independent of the stage of disease, since no benefit has yet been observed among patients with stage I disease?
Fabiana Cecere, M.D.
Emilio Bria, M.D.
Regina Elena National Cancer Institute
00144 Rome, Italy
Rafael Rosell, M.D.
Catalan Institute of Oncology
08916 Barcelona, Spain
rrosell@ico.scs.es
References
The International Adjuvant Lung Cancer Trial Collaborative Group. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N Engl J Med 2004;350:351-360.
To the Editor: In the report by Olaussen et al., the data were not stratified according to age, sex, smoking status, or histologic type. Will a significant survival benefit persist after adjustment for all these factors?
Tal Grenader, M.D.
Linda Shavit, M.D.
Shaare Zedek Medical Center
91031 Jerusalem, Israel
talgrenader65@hotmail.com
The authors reply: Our analyses addressed, in three steps, the question of whether the characteristics of the tumors or of the patients influenced the observed differential effect of chemotherapy on survival among patients with ERCC1-negative tumors and those with ERCC1-positive tumors. First, we verified that there was no significant interaction between the treatment effect and any of the demographic and clinical characteristics of patients included in the IALT Biology (IALT Bio) study, a subgroup of the IALT study population.1 This step was also taken in the analyses performed in the IALT. The absence of an interaction was verified in particular for such factors as histologic type, stage of disease, and the assigned treatment (P=0.64, P=0.45, and P=0.26, respectively). Second, when studying the interaction between ERCC1 status and treatment, we adjusted not only for stratification and prognostic factors but also for factors associated with ERCC1 status (histopathological type, the presence or absence of pleural invasion, and age). Third, we verified that the interaction with ERCC1 status would not disappear when any other interaction was added to the Cox model (P values for an interaction with ERCC1 status remained below 0.009 in the presence of an interaction with histologic type, stage of disease, or assigned treatment). Thus, the interaction between the ERCC1 status and treatment was robust enough to withstand the addition of any other interaction.
Wilcox points out the different histologic types of carcinoma among the patients with ERCC1-negative and ERCC1-positive tumors (P<0.001, in Table 1 of our article). As shown above, this difference does not explain the interaction between ERCC1 status and the treatment effect.
Cecere and colleagues request clarification of the role of the assigned drug and the stage of disease in the interaction between ERCC1 status and treatment. Our analysis shows that this interaction is the only one that remains significant when the model includes either of these two interactions (drug or stage of disease).
Grenader and Shavit are concerned that stratification according to age, sex, smoking status, and histologic type was not performed. Our analysis took all these factors into account, except smoking status. Data on smoking status were not collected in the trial. Smoking may affect the prognosis for patients with lung cancer and may be related to ERCC1 expression. However, since the proportion of smokers is quite high among patients with non–small-cell lung cancer (around 90%), only a very wide variation in the treatment effect between smokers and nonsmokers could explain the observed significant interaction between treatment and ERCC1 status. More important, adjustments for histologic type and sex (which are classically associated with smoking) and the corresponding interactions with treatment do not change the results.
Ariane Dunant, M.S.
Pierre Fouret, M.D., Ph.D.
Jean-Charles Soria, M.D., Ph.D.
Institut de Cancérologie Gustave Roussy
94805 Villejuif CEDEX, France
soria@igr.fr
References
The International Adjuvant Lung Cancer Trial Collaborative Group. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N Engl J Med 2004;350:351-360.