Letrozole in Breast Cancer
http://www.100md.com
《新英格兰医药杂志》
To the Editor: Goss et al., who conducted a trial of letrozole in postmenopausal women after five years of tamoxifen therapy for breast cancer (Nov. 6 issue),1 present their results in a manner that could influence treatment decisions more strongly than is merited by the data provided. The focus on estimates of disease-free survival four years after tamoxifen therapy, when less than 1 percent of the patients had been followed at least four years, is potentially misleading. The preliminary survival data do not show a treatment difference (95 percent confidence interval, 0.48 to 1.21). It is inappropriate that the discussion of the survival results suggests otherwise. In addition, the absence of more than two years of safety data overall, and of any safety data regarding fractures, osteoporosis, and cardiovascular events for 834 patients, is especially of concern.
Although they represent an advance for women who complete five years of tamoxifen therapy, these data need careful interpretation because of the limitations imposed by their preliminary nature.
Aman U. Buzdar, M.D.
University of Texas M.D. Anderson Cancer Center
Houston, TX 77030
abuzdar@mdanderson.org
References
Goss PE, Ingle JN, Martino S, et al. A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med 2003;349:1793-1802.
To the Editor: The study of letrozole by Goss et al. is important, but as Burstein1 and Bryant and Wolmark2 state in their editorials, we do not yet know which patients should receive letrozole and for how long. Letrozole will benefit only some of the 2 to 3 percent of patients with breast cancer each year who have a relapse after five years of follow-up.
We suggest that the authors clarify effect sizes in a manner that can be understood by doctors and patients.3 There were 2.2 percent fewer relapses or new cancers in the letrozole group than in the placebo group at 2.4 years. This means 22 fewer cancers per 1000 women treated, which translates into treatment for 45 women for 2.4 years (or 109 women for 1 year) to prevent 1 cancer. The effect of treatment will probably diminish in subsequent years. The cost of this treatment will be 21 cases of arthritis, 6 fractures, 5 cardiovascular events, and other side effects. As yet, no significant effect on mortality has been documented.
P?ivi Hietanen, M.D., Ph.D.
Finnish Medical Journal
00500 Helsinki, Finland
Marjukka M?kel?, M.D., Ph.D.
Finnish Office for Health Technology
00530 Helsinki, Finland
References
Burstein HJ. Beyond tamoxifen -- extending endocrine treatment for early-stage breast cancer. N Engl J Med 2003;349:1857-1859.
Bryant J, Wolmark N. Letrozole after tamoxifen for breast cancer: what is the price of success? N Engl J Med 2003;349:1855-1857.
Gigerenzer G, Edwards A. Simple tools for understanding risks: from innumeracy to insight. BMJ 2003;327:741-744.
To the Editor: Given the results of the trial conducted by Goss et al., oncologists are now obliged to consider therapy with an aromatase inhibitor after five years of tamoxifen therapy in postmenopausal patients with estrogen-receptor–positive breast cancer. The trial was prematurely halted, and the authors acknowledged that the optimal duration of therapy remains to be determined. It is unknown whether a longer duration of treatment with letrozole could be detrimental, as was the case with tamoxifen.1
In practice, many postmenopausal patients with estrogen-receptor–positive breast cancer cannot take tamoxifen because of intolerance or thromboembolism. Since the efficacy of aromatase inhibitors is equal to that of tamoxifen,2 it is common practice to replace tamoxifen with an aromatase inhibitor and complete a five-year course of adjuvant therapy. Now, given the information presented by Goss et al., clinicians will face a dilemma concerning the total duration of therapy in patients who take an aromatase inhibitor as their primary adjuvant hormonal treatment. They will have to decide whether to complete a 10-year adjuvant course or to discontinue therapy at 5 years, since the long-term efficacy and safety profile (especially with respect to bone) are not known for this class of drug.
Nasir Shahab, M.D.
Ellis Fischel Cancer Center
Columbia, MO 65203
References
Fisher B, Digman J, Bryant J, Wolmark N. Five versus more than five years of tamoxifen for lymph node-negative breast cancer: updated findings from the National Surgical Adjuvant Breast and Bowel Project B-14 randomized trial. J Natl Cancer Inst 2001;93:684-690.
Baum M, Budzar AU, Cuzick J, et al. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial. Lancet 2002;359:2131-2139.
To the Editor: As one of the "hundreds of thousands of women"1 finishing tamoxifen this year, I read the report by Goss and colleagues and the accompanying editorials by Burstein and by Bryant and Wolmark with great interest. Since I am someone with "greater residual jeopardy for recurrence,"1 the potential benefits of letrozole for me are particularly promising.
As a clinical researcher, I also understand that the long-term safety of this drug is, as yet, unknown and may include increased risks of osteoporosis and cardiac events. Speaking, however, as one of those hundreds of thousands of women who are soon to have the safety net of tamoxifen withdrawn, I can say that my fear of a recurrence of breast cancer, especially distant metastasis, far outweighs any fears of cardiovascular disease or diminished bone mineral density. Although there are lifestyle changes and effective drugs to enhance bone density and to minimize the risk of cardiovascular disease, there is little we can do to prevent recurrent breast cancer.
Susan R. Harris, Ph.D.
University of British Columbia
Vancouver, BC V6T 2B5, Canada
References
Burstein HJ. Beyond tamoxifen -- extending endocrine treatment for early-stage breast cancer. N Engl J Med 2003;349:1857-1859.
To the Editor: Bryant and Wolmark suggest two ways to "avoid" early cessation of clinical trials: first, trials should be stopped only if found unsafe, and second, a minimal follow-up must be completed before early reporting of improved efficacy is permitted. These suggestions reflect the decision to weigh society's interest in gathering reliable information more heavily than the interests of the women enrolled in the study. Specifying this choice in advance merely shifts the decision-making authority from the physicians conducting the study to a higher level.
Whereas avoiding harm may be the physician's most basic obligation, the primary reason a patient consults a physician is to obtain a benefit. Similarly, specifying a minimal time before evaluation of efficacy is allowed hides information from the physician. This example demonstrates that randomized clinical trials, even when ethically acceptable initially, are fraught with continuing potential conflicts.1 Because care of the patient is the doctor's primary responsibility, the modifications of this duty suggested by Bryant and Wolmark bear the burden of justification — which the editorialists do not provide.
Samuel Hellman, M.D.
University of Chicago
Chicago, IL 60637
s-hellman@uchicago.edu
Deborah Hellman, J.D.
University of Maryland
Baltimore, MD 21201
References
Hellman S, Hellman DS. Of mice but not men: problems of the randomized clinical trial. N Engl J Med 1991;324:1585-1589.
The authors reply: In response to Dr. Buzdar and Drs. Hietanen and M?kel?: in Table 4 of our article we present absolute differences in disease-free survival between the study groups, with 95 percent confidence intervals, for years 1 to 4. The 5 percent reduction in disease-free survival at three years means that 20 patients need to be treated for three years to prevent one recurrence. The absolute benefit for a given patient depends both on the base-line risk of recurrence and on the proportional reduction in the rate of recurrence,1 which was 43 percent in our trial. This compares well with the 28 percent and 50 percent improvements seen in patients with estrogen-receptor–positive cancer who were given tamoxifen for two and five years, respectively.2 Our absolute improvement of 5 percent at three years also compares favorably with the advantage in disease-free survival of 1.7 percent seen at three years with anastrozole in the recent ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial.3
We disagree with Dr. Shahab that our trial was discontinued prematurely. We sought a reduction in risk (22 percent) that might change clinical practice. Had the drug been only as effective as we hypothesized, proof would have required six years (four years of accrual and two years of follow-up). Instead, letrozole exceeded expectations, yielding a result (a 43 percent reduction) at the first interim analysis after five years. We never planned to treat all the patients for five years before analyzing our results. A planned rerandomization and follow-up of our patients for an additional five years will help to resolve the issue of the optimal duration of letrozole treatment and to address concerns about its long-term safety. Meanwhile, we believe our trial has provided definitive data on an important clinical end point that can be used now in clinical decision making.
We are pleased that Dr. Harris agrees with our opinion that preventing the recurrence of disease in patients with early breast cancer is itself a laudable goal. The perspective of a patient in this regard provides a powerful statement that should be recognized and appreciated.
The editorials accompanying our article raise questions about the primary end point, the omission of overall survival as a primary end point, the magnitude of the benefit associated with the use of letrozole in terms of the number needed to treat, and the optimal duration of letrozole treatment.
Contrary to Bryant and Wolmark's comments, the only planned primary end point in our trial was disease-free survival, as stated in our article. There is strong precedent for the use of disease-free survival as the single primary end point in trials of adjuvant hormonal therapy for breast cancer. The Food and Drug Administration approved tamoxifen and anastrozole for use as adjuvant therapy on the basis of data on disease-free survival.4 In addition, an overview by the Early Breast Cancer Trialists' Collaborative Group5 indicates that differences in disease-free survival regularly predict differences in overall survival. We think it would have been unethical to await a survival advantage in the presence of a statistically and clinically significant difference in disease-free survival.
With regard to the magnitude of the difference in disease-free survival, the number needed to treat with letrozole noted by Burstein in his editorial — 100 patients per year to prevent one event — is correct but misleading because it fails to account for long-term administration. The number needed to treat when letrozole is given for three years (20 patients to prevent one recurrence) compares favorably, for example, with the number of patients with hypertension who need to be treated for five years to prevent one cardiovascular event (e.g., myocardial infarction, death, or stroke).6
Paul E. Goss, M.D., Ph.D.
Princess Margaret Hospital
Toronto, ON M5G 2M9, Canada
pegoss@interlog.com
James N. Ingle, M.D.
Mayo Clinic
Rochester, MN 55905
Joseph L. Pater, M.D.
Queen's University
Kingston, ON K7L 3N6, Canada
for the Principal Investigators
References
Wieand HS. Is relative risk reduction a useful measure for patients or families who must choose a method of treatment? J Clin Oncol 2003;21:4263-4264.
Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet 1998;351:1451-1467.
Baum M, Budzar AU, Cuzick J, et al. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial. Lancet 2002;359:2131-2139.
Johnson JR, Williams G, Pazdur R. End points and United States Food and Drug Administration approval of oncology drugs. J Clin Oncol 2003;21:1404-1411.
Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet 1998;351:1451-1467.
Gueyffier F, Boutitie F, Boissel JP, et al. Effect of hypertensive drug treatment on cardiovascular outcomes in women and men: a meta-analysis of individual patient data from randomized, controlled trials. Ann Intern Med 1997;126:761-767.
Drs. Bryant and Wolmark reply: We do not disagree with Goss et al. that the primary end point in their trial was disease-free survival. Indeed, in our editorial, we state that "the interim analysis was carried out strictly according to protocol. . . . Furthermore, the primary end point (the time to a first recurrence or new contralateral breast cancer), the number of events required to trigger the interim analysis, and the specific statistical tests used in the interim analysis were all specified a priori." However, the primary objective as stated in the protocol (section 1.1) was "to determine the DFS and overall survival for women who have previously received 5 years of adjuvant tamoxifen, randomized to receive either letrozole 2.5 mg daily or placebo daily for 5 years."1 Thus, the argument that survival was not an important consideration is unconvincing, and because of crossover it may now be questionable whether documentation of a survival benefit will be forthcoming.
We believe that Hellman and Hellman overly simplify our comments. In fact, our first suggestion was to consider "recommending that the data and safety monitoring board should not generally stop a trial early except for reasons of safety if doing so would compromise a primary aim of the trial." Clearly, the qualifier here is important. This point gets to the heart of our comment above that, even though it was not a stated primary end point, eventual confirmation of a survival benefit was nevertheless a goal of this study.
John Bryant, Ph.D.
University of Pittsburgh
Pittsburgh, PA 15260
Norman Wolmark, M.D.
Allegheny General Hospital
Pittsburgh, PA 15212
References
A phase III randomized double-blind study of letrozole versus placebo in women with primary breast cancer completing five or more years of tamoxifen. Amendment. Kingston, Ont., Canada: National Cancer Institute of Canada Clinical Trials Group, November 14, 2001.
Dr. Burstein replies: Expressing study benefits in terms of events per year helps individual doctors and patients make better medical decisions. The calculation of the number to treat or the number of events per year assumes that the likelihood of events is uniform throughout the patient population and over time — conditions that may not have been present in the trial reported by Goss et al. For instance, women with different base-line levels of the risk of tumor recurrence (e.g., patients with node-positive disease vs. those with node-negative disease) receive different absolute benefits even with the same relative risk reduction. Similarly, as the risk of recurrence declines over time, the likely benefits of ongoing letrozole treatment may diminish correspondingly. Longer follow-up, beyond the reported median of 2.4 years, will be needed to establish the effects of long-term administration.
Harold J. Burstein, M.D., Ph.D.
Dana–Farber Cancer Institute
Boston, MA 02115
hburstein@partners.org
Although they represent an advance for women who complete five years of tamoxifen therapy, these data need careful interpretation because of the limitations imposed by their preliminary nature.
Aman U. Buzdar, M.D.
University of Texas M.D. Anderson Cancer Center
Houston, TX 77030
abuzdar@mdanderson.org
References
Goss PE, Ingle JN, Martino S, et al. A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med 2003;349:1793-1802.
To the Editor: The study of letrozole by Goss et al. is important, but as Burstein1 and Bryant and Wolmark2 state in their editorials, we do not yet know which patients should receive letrozole and for how long. Letrozole will benefit only some of the 2 to 3 percent of patients with breast cancer each year who have a relapse after five years of follow-up.
We suggest that the authors clarify effect sizes in a manner that can be understood by doctors and patients.3 There were 2.2 percent fewer relapses or new cancers in the letrozole group than in the placebo group at 2.4 years. This means 22 fewer cancers per 1000 women treated, which translates into treatment for 45 women for 2.4 years (or 109 women for 1 year) to prevent 1 cancer. The effect of treatment will probably diminish in subsequent years. The cost of this treatment will be 21 cases of arthritis, 6 fractures, 5 cardiovascular events, and other side effects. As yet, no significant effect on mortality has been documented.
P?ivi Hietanen, M.D., Ph.D.
Finnish Medical Journal
00500 Helsinki, Finland
Marjukka M?kel?, M.D., Ph.D.
Finnish Office for Health Technology
00530 Helsinki, Finland
References
Burstein HJ. Beyond tamoxifen -- extending endocrine treatment for early-stage breast cancer. N Engl J Med 2003;349:1857-1859.
Bryant J, Wolmark N. Letrozole after tamoxifen for breast cancer: what is the price of success? N Engl J Med 2003;349:1855-1857.
Gigerenzer G, Edwards A. Simple tools for understanding risks: from innumeracy to insight. BMJ 2003;327:741-744.
To the Editor: Given the results of the trial conducted by Goss et al., oncologists are now obliged to consider therapy with an aromatase inhibitor after five years of tamoxifen therapy in postmenopausal patients with estrogen-receptor–positive breast cancer. The trial was prematurely halted, and the authors acknowledged that the optimal duration of therapy remains to be determined. It is unknown whether a longer duration of treatment with letrozole could be detrimental, as was the case with tamoxifen.1
In practice, many postmenopausal patients with estrogen-receptor–positive breast cancer cannot take tamoxifen because of intolerance or thromboembolism. Since the efficacy of aromatase inhibitors is equal to that of tamoxifen,2 it is common practice to replace tamoxifen with an aromatase inhibitor and complete a five-year course of adjuvant therapy. Now, given the information presented by Goss et al., clinicians will face a dilemma concerning the total duration of therapy in patients who take an aromatase inhibitor as their primary adjuvant hormonal treatment. They will have to decide whether to complete a 10-year adjuvant course or to discontinue therapy at 5 years, since the long-term efficacy and safety profile (especially with respect to bone) are not known for this class of drug.
Nasir Shahab, M.D.
Ellis Fischel Cancer Center
Columbia, MO 65203
References
Fisher B, Digman J, Bryant J, Wolmark N. Five versus more than five years of tamoxifen for lymph node-negative breast cancer: updated findings from the National Surgical Adjuvant Breast and Bowel Project B-14 randomized trial. J Natl Cancer Inst 2001;93:684-690.
Baum M, Budzar AU, Cuzick J, et al. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial. Lancet 2002;359:2131-2139.
To the Editor: As one of the "hundreds of thousands of women"1 finishing tamoxifen this year, I read the report by Goss and colleagues and the accompanying editorials by Burstein and by Bryant and Wolmark with great interest. Since I am someone with "greater residual jeopardy for recurrence,"1 the potential benefits of letrozole for me are particularly promising.
As a clinical researcher, I also understand that the long-term safety of this drug is, as yet, unknown and may include increased risks of osteoporosis and cardiac events. Speaking, however, as one of those hundreds of thousands of women who are soon to have the safety net of tamoxifen withdrawn, I can say that my fear of a recurrence of breast cancer, especially distant metastasis, far outweighs any fears of cardiovascular disease or diminished bone mineral density. Although there are lifestyle changes and effective drugs to enhance bone density and to minimize the risk of cardiovascular disease, there is little we can do to prevent recurrent breast cancer.
Susan R. Harris, Ph.D.
University of British Columbia
Vancouver, BC V6T 2B5, Canada
References
Burstein HJ. Beyond tamoxifen -- extending endocrine treatment for early-stage breast cancer. N Engl J Med 2003;349:1857-1859.
To the Editor: Bryant and Wolmark suggest two ways to "avoid" early cessation of clinical trials: first, trials should be stopped only if found unsafe, and second, a minimal follow-up must be completed before early reporting of improved efficacy is permitted. These suggestions reflect the decision to weigh society's interest in gathering reliable information more heavily than the interests of the women enrolled in the study. Specifying this choice in advance merely shifts the decision-making authority from the physicians conducting the study to a higher level.
Whereas avoiding harm may be the physician's most basic obligation, the primary reason a patient consults a physician is to obtain a benefit. Similarly, specifying a minimal time before evaluation of efficacy is allowed hides information from the physician. This example demonstrates that randomized clinical trials, even when ethically acceptable initially, are fraught with continuing potential conflicts.1 Because care of the patient is the doctor's primary responsibility, the modifications of this duty suggested by Bryant and Wolmark bear the burden of justification — which the editorialists do not provide.
Samuel Hellman, M.D.
University of Chicago
Chicago, IL 60637
s-hellman@uchicago.edu
Deborah Hellman, J.D.
University of Maryland
Baltimore, MD 21201
References
Hellman S, Hellman DS. Of mice but not men: problems of the randomized clinical trial. N Engl J Med 1991;324:1585-1589.
The authors reply: In response to Dr. Buzdar and Drs. Hietanen and M?kel?: in Table 4 of our article we present absolute differences in disease-free survival between the study groups, with 95 percent confidence intervals, for years 1 to 4. The 5 percent reduction in disease-free survival at three years means that 20 patients need to be treated for three years to prevent one recurrence. The absolute benefit for a given patient depends both on the base-line risk of recurrence and on the proportional reduction in the rate of recurrence,1 which was 43 percent in our trial. This compares well with the 28 percent and 50 percent improvements seen in patients with estrogen-receptor–positive cancer who were given tamoxifen for two and five years, respectively.2 Our absolute improvement of 5 percent at three years also compares favorably with the advantage in disease-free survival of 1.7 percent seen at three years with anastrozole in the recent ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial.3
We disagree with Dr. Shahab that our trial was discontinued prematurely. We sought a reduction in risk (22 percent) that might change clinical practice. Had the drug been only as effective as we hypothesized, proof would have required six years (four years of accrual and two years of follow-up). Instead, letrozole exceeded expectations, yielding a result (a 43 percent reduction) at the first interim analysis after five years. We never planned to treat all the patients for five years before analyzing our results. A planned rerandomization and follow-up of our patients for an additional five years will help to resolve the issue of the optimal duration of letrozole treatment and to address concerns about its long-term safety. Meanwhile, we believe our trial has provided definitive data on an important clinical end point that can be used now in clinical decision making.
We are pleased that Dr. Harris agrees with our opinion that preventing the recurrence of disease in patients with early breast cancer is itself a laudable goal. The perspective of a patient in this regard provides a powerful statement that should be recognized and appreciated.
The editorials accompanying our article raise questions about the primary end point, the omission of overall survival as a primary end point, the magnitude of the benefit associated with the use of letrozole in terms of the number needed to treat, and the optimal duration of letrozole treatment.
Contrary to Bryant and Wolmark's comments, the only planned primary end point in our trial was disease-free survival, as stated in our article. There is strong precedent for the use of disease-free survival as the single primary end point in trials of adjuvant hormonal therapy for breast cancer. The Food and Drug Administration approved tamoxifen and anastrozole for use as adjuvant therapy on the basis of data on disease-free survival.4 In addition, an overview by the Early Breast Cancer Trialists' Collaborative Group5 indicates that differences in disease-free survival regularly predict differences in overall survival. We think it would have been unethical to await a survival advantage in the presence of a statistically and clinically significant difference in disease-free survival.
With regard to the magnitude of the difference in disease-free survival, the number needed to treat with letrozole noted by Burstein in his editorial — 100 patients per year to prevent one event — is correct but misleading because it fails to account for long-term administration. The number needed to treat when letrozole is given for three years (20 patients to prevent one recurrence) compares favorably, for example, with the number of patients with hypertension who need to be treated for five years to prevent one cardiovascular event (e.g., myocardial infarction, death, or stroke).6
Paul E. Goss, M.D., Ph.D.
Princess Margaret Hospital
Toronto, ON M5G 2M9, Canada
pegoss@interlog.com
James N. Ingle, M.D.
Mayo Clinic
Rochester, MN 55905
Joseph L. Pater, M.D.
Queen's University
Kingston, ON K7L 3N6, Canada
for the Principal Investigators
References
Wieand HS. Is relative risk reduction a useful measure for patients or families who must choose a method of treatment? J Clin Oncol 2003;21:4263-4264.
Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet 1998;351:1451-1467.
Baum M, Budzar AU, Cuzick J, et al. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial. Lancet 2002;359:2131-2139.
Johnson JR, Williams G, Pazdur R. End points and United States Food and Drug Administration approval of oncology drugs. J Clin Oncol 2003;21:1404-1411.
Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet 1998;351:1451-1467.
Gueyffier F, Boutitie F, Boissel JP, et al. Effect of hypertensive drug treatment on cardiovascular outcomes in women and men: a meta-analysis of individual patient data from randomized, controlled trials. Ann Intern Med 1997;126:761-767.
Drs. Bryant and Wolmark reply: We do not disagree with Goss et al. that the primary end point in their trial was disease-free survival. Indeed, in our editorial, we state that "the interim analysis was carried out strictly according to protocol. . . . Furthermore, the primary end point (the time to a first recurrence or new contralateral breast cancer), the number of events required to trigger the interim analysis, and the specific statistical tests used in the interim analysis were all specified a priori." However, the primary objective as stated in the protocol (section 1.1) was "to determine the DFS and overall survival for women who have previously received 5 years of adjuvant tamoxifen, randomized to receive either letrozole 2.5 mg daily or placebo daily for 5 years."1 Thus, the argument that survival was not an important consideration is unconvincing, and because of crossover it may now be questionable whether documentation of a survival benefit will be forthcoming.
We believe that Hellman and Hellman overly simplify our comments. In fact, our first suggestion was to consider "recommending that the data and safety monitoring board should not generally stop a trial early except for reasons of safety if doing so would compromise a primary aim of the trial." Clearly, the qualifier here is important. This point gets to the heart of our comment above that, even though it was not a stated primary end point, eventual confirmation of a survival benefit was nevertheless a goal of this study.
John Bryant, Ph.D.
University of Pittsburgh
Pittsburgh, PA 15260
Norman Wolmark, M.D.
Allegheny General Hospital
Pittsburgh, PA 15212
References
A phase III randomized double-blind study of letrozole versus placebo in women with primary breast cancer completing five or more years of tamoxifen. Amendment. Kingston, Ont., Canada: National Cancer Institute of Canada Clinical Trials Group, November 14, 2001.
Dr. Burstein replies: Expressing study benefits in terms of events per year helps individual doctors and patients make better medical decisions. The calculation of the number to treat or the number of events per year assumes that the likelihood of events is uniform throughout the patient population and over time — conditions that may not have been present in the trial reported by Goss et al. For instance, women with different base-line levels of the risk of tumor recurrence (e.g., patients with node-positive disease vs. those with node-negative disease) receive different absolute benefits even with the same relative risk reduction. Similarly, as the risk of recurrence declines over time, the likely benefits of ongoing letrozole treatment may diminish correspondingly. Longer follow-up, beyond the reported median of 2.4 years, will be needed to establish the effects of long-term administration.
Harold J. Burstein, M.D., Ph.D.
Dana–Farber Cancer Institute
Boston, MA 02115
hburstein@partners.org