Correspondence
Adjuvant Therapy for Breast Cancer
N Engl J Med 1994; 331:741-746September 15, 1994
- Article
To the Editor:
Wood et al. (May 5 issue)1 report the results of a randomized study of different doses and levels of dose intensity (dose per unit of time) of adjuvant chemotherapy in women with stage II, node-positive breast cancer. The study is an excellent example of the merits of the conventional study design, but also of its limitations.
Wood et al. correctly conclude that the toxicity of treatment is directly related to the dose intensity, the dose rate, or both (their Table 1)1. This means that the toxicity is related to the area under the concentration-time curve after a single dose, the peak concentration of the three study drugs, or both. In this particular trial such a conclusion could be drawn without the determination of plasma concentrations.
However, for therapeutic effects the situation is different. The survival data showed no difference in outcome between group 1 (the high dose rate) and group 2 (the medium dose rate), whereas the outcome of group 3 (the low dose rate) was clearly inferior. Because groups 1 and 2 received identical cumulative doses, it is not clear whether the effect on survival was correlated with the dose rate or the cumulative dose. In order to gain as much information as possible from such trials, individual pharmacokinetic-pharmacodynamic approaches have been proposed2. Analysis of plasma samples taken at random from every patient3 would allow the relation between individual outcomes and pharmacokinetic measures to be explored. Individual pharmacokinetic measures can be estimated from a few plasma concentrations per patient with use of a population approach4.
Variability among individuals gives a wide range of measures and outcomes, allowing tests of the influence of the dose rate (represented by the peak concentration of the three drugs), the cumulative dose (represented by the cumulative area under the time-concentration curve), or the area under the curve above a certain threshold concentration (as a combined measure of concentration and duration of exposure) on therapeutic effect and toxicity. Thus, it is possible to evaluate the risk-benefit ratios of different dose schedules, as has been discussed recently5. If this information were available, we doubt whether the authors' conclusion (“the doses of chemotherapy . . . should not be reduced if the maximal benefit is to be achieved”) would hold true.
5 ReferencesHans-Joachim Muller, M.D.
Christoph H. Gleiter, M.D.
Ursula Gundert-Remy, M.D.
University Hospital, 37075 Gottingen, Germany1
Wood WC ,Budman DR ,Korzun AH , et al. Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N Engl J Med 1994;330:1253-1259
Full Text | Web of Science | Medline2
Peck CC. Simultaneous pharmacokinetic-pharmacodynamic modelling in drug development and registration. Presented at Symposium on Advances in Simultaneous Pharmacokinetic-Pharmacodynamic Modelling, Noordwijk, the Netherlands, June 28-30, 1990. abstract.
3
Sheiner LB ,Benet LZ . Premarketing observational studies of population pharmacokinetics of new drugs. Clin Pharmacol Ther 1985;38:481-487
CrossRef | Web of Science | Medline4
Sheiner LB ,Rosenberg B ,Marathe VV . Estimation of population characteristics of pharmacokinetic parameters from routine clinical data. J Pharmacokinet Biopharm 1977;5:445-479
CrossRef | Medline5
Baber N . International conference on harmonisation of technical requirements for registration of pharmaceuticals for human use (ICH). Br J Clin Pharmacol 1994;37:401-404
Web of Science | Medline
To the Editor:
With a median follow-up of 3.4 years, the results of Wood et al. are still relatively early and may therefore be subject to change. Previous work in this field has indicated that this amount of follow-up may not be sufficient for firm conclusions to be drawn. A good example is the original study of adjuvant therapy with cyclophosphamide, methotrexate, and fluorouracil (CMF) in stage II breast cancer by Bonadonna et al. In their first study, which had a median follow-up of 27 months, postmenopausal women with stage II breast cancer treated with CMF had a survival advantage as compared with controls1. However, with further follow-up this survival advantage was no longer present2.
It is difficult to say what the optimal length of follow-up should be in order to draw conclusions about policy, but given the natural history of breast cancer and previous experience with clinical trials in patients with this illness, a median follow-up of 3.4 years is probably not sufficient. Therefore, the results presented should be regarded as preliminary. As a result, policy statements about dose intensity and the treatment of postmenopausal women should be deferred until further follow-up data have been obtained.
2 ReferencesDavid Melnychuk, M.D.
Lawrence C. Panasci, M.D.
Jewish General Hospital, Montreal, QC H3T 1E2, Canada1
Bonadonna G ,Brusamolino E ,Valagussa P , et al. Combination chemotherapy as an adjuvant treatment in operable breast cancer. N Engl J Med 1976;294:405-410
Full Text | Web of Science | Medline2
Rossi A ,Bonadonna G ,Valagussa P ,Veronesi V . Multimodal treatment in operable breast cancer: five-year results of the CMF programme. BMJ 1981;282:1427-1431
CrossRef | Web of Science | Medline
To the Editor:
The design of the trial of adjuvant chemotherapy for breast cancer conducted by the Cancer and Leukemia Group B1 unfortunately does not isolate any of the three main chemotherapy variables -- dose size, dose intensity, and total dose2. It is notable, however, that despite different dose intensities there is no difference in the outcome of the two groups assigned to receive the same total dose, whereas there is a significantly inferior outcome among the patients randomly assigned to receive the low total dose. One may conclude that the total dose is the underlying basis for the observed differences, whereas the authors prefer to postulate a nonlinear dose-response or intensity-response effect.
If the survival benefit of adjuvant chemotherapy mostly accrues to patients who are incurable3 because of a drug-resistant minority clone, then both theory4 and laboratory models5 suggest that total dose is indeed the important determinant up to the point at which the sensitive cell population is eliminated in the majority of patients. If so, the conclusion that doses “should not be reduced” is unsound and may lead to inappropriate risk taking in the face of excessive toxicity. Although it is good clinical practice to give the maximal well-tolerated dose, it may be possible to compensate for dose reductions or delays with additional therapy to achieve the target total dose.
5 ReferencesChris M.L. Coppin, B.M., D.Phil.
James H. Goldie, M.D.
British Columbia Cancer Agency, Vancouver, BC V5Z 4E6, Canada1
Wood WC ,Budman DR ,Korzun AH , et al. Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N Engl J Med 1994;330:1253-1259
Full Text | Web of Science | Medline2
Coppin CML . The description of chemotherapy delivery: options and pitfalls. Semin Oncol 1987;14:Suppl 4:34-42
Web of Science | Medline3
Clark GM ,Osborne CK . Prognostic factors in breast cancer. N Engl J Med 1992;327:1318-1318
Full Text | Web of Science4
Coppin C ,Coldman A ,Goldie J . Modelling the effect of chemotherapy intensity. Prog Clin Biol Res 1990;354:103-1165
Skipper HE. Dose intensity versus total dose of chemotherapy: an experimental basis. In: DeVita VT Jr, Hellman S, Rosenberg SA, eds. Important advances in oncology 1990. Philadelphia: J.B. Lippincott, 1990:43-64.
To the Editor:
In the study of the dose and dose intensity of adjuvant chemotherapy for breast cancer recently published in the Journal, one patient died three years after being treated for cardiomyopathy1. Given the well-known toxic effects of doxorubicin on the heart, doxorubicin-free combinations should be carefully evaluated. CMF chemotherapy is equally effective, at least in patients with fewer than four positive axillary lymph nodes2-5. It is unfortunate that many clinical trials use doxorubicin as adjuvant chemotherapy for all patients with node-positive disease, regardless of risk factors. In a large proportion of young women with breast cancer, the implications of asymptomatic cardiomyopathy may not be fully realized for many years. I hope that local ethics committees will interfere and that for the time being they will prevent patients with fewer than four positive axillary lymph nodes from entering trials that offer only adjuvant chemotherapy including doxorubicin.
5 ReferencesChristian Sauter, M.D.
University Hospital, CH-8091 Zurich, Switzerland1
Wood WC ,Budman DR ,Korzun AH , et al. Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N Engl J Med 1994;330:1253-1259
Full Text | Web of Science | Medline2
Carpenter JT ,Velez-Garcia E ,Aron BS , et al. Prospective randomized comparison of cyclophosphamide, doxorubicin (adriamycin) and fluorouracil (CAF) vs. cyclophosphamide, methotrexate and fluorouracil (CMF) for breast cancer with positive axillary nodes: a Southeastern Cancer Study Group study. Prog Proc Am Soc Clin Oncol 1991;10:45-45 abstract.3
O'Bryan R ,Green S ,O'Sullivan J , et al. A comparison of CMFVP for one year to short term adriamycin based chemotherapy for patients with receptor-negative node positive operable breast cancer: an inter-group study. Prog Proc Am Soc Clin Oncol 1992;11:61-61 abstract.4
Fisher B ,Brown AM ,Dimitrov NV , et al. Two months of doxorubicin-cyclophosphamide with and without interval reinduction therapy compared with 6 months of cyclophosphamide, methotrexate, and fluorouracil in positive-node breast cancer patients with tamoxifen-nonresponsive tumors: results from the National Surgical Adjuvant Breast and Bowel Project B-15. J Clin Oncol 1990;8:1483-1496
Web of Science | Medline5
Fisher B, Costantino J, Redmond C, Wickerham DL, NSABP Investigators. Recent information from current NSABP trials of adjuvant therapy for breast cancer. In: Salmon SE, ed. Adjuvant therapy of cancer VII. Philadelphia: J.B. Lippincott, 1993:148-61.
To the Editor:
The study by Wood et al. evaluating the response of breast cancer in relation to the chemotherapeutic dose and dose intensity used the standard prognostic factors of tumor size and node involvement, along with the estrogen- and progesterone-receptor content of the tumor, to evaluate outcome1. We now report that patients with a family history of breast cancer have a shorter time to recurrence than patients with no such history.
We evaluated 137 women receiving radiation therapy for breast cancer between 1984 and 1994. Patients were selected if data were available on their tumor size, nodal involvement, any family history of breast cancer, and survival status. Information on the family history of breast cancer was obtained from the patient's chart or by interview. Tumor size and axillary-node status were determined from the surgical pathology report. The patient was considered to have no axillary-node involvement if histologic examination showed no node involvement after axillary dissection. The axillary nodes were considered to be involved if such involvement was demonstrated by histopathological examination. Only data on patients with infiltrating tumors were included. Statistical analyses were performed with the SPSS system2.
The 41 women with a family history of breast cancer had a significantly shorter time to recurrence than the 96 women without a family history (P = 0.02 by the log-rank test) (Figure 1Figure 1
Kaplan-Meier Plot of Cumulative Recurrence-free Survival among 137 Women with Breast Cancer.). But there was no significant difference in the time to recurrence between 20 women with an affected first-degree relative (a mother or sister) and 21 women with a more distant affected relative (a grandmother, cousin, aunt, or niece) (P = 0.7 by the log-rank test) (Figure 1). Proportional-hazards regression demonstrated that a family history of breast cancer has a significant effect on time to recurrence, independently of tumor size and nodal involvement (Table 1Table 1
Cox Proportional-Hazards Regression Analysis of Cumulative Recurrence-free Survival among 140 Patients with Breast Cancer, with Control for Family History of Breast Cancer, Tumor Size, and Presence or Absence of Nodal Involvement.).Because of the need to identify patients with breast cancer who can benefit from adjuvant therapy and to spare others the side effects, much effort has been spent testing new prognostic indicators3. Family history of breast cancer, which is an easily obtained prognostic factor, may complement those already in use. Furthermore, it may be the only risk factor that is also a prognostic factor, since there is no association between survival and reproductive or hormonal risk factors, dietary variables, alcohol consumption, or smoking4. Additional studies would be worthwhile in order to determine whether family history is indeed an independent predictor or whether it may be correlated with other predictors already in use.
4 ReferencesJoan Garey, M.A.
Steven Lehrer, M.D.
Mount Sinai School of Medicine, New York, NY 100291
Wood WC ,Budman DR ,Korzun AH , et al. Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N Engl J Med 1994;330:1253-1259
Full Text | Web of Science | Medline2
Norusis MJ. SPSS for Windows: advanced statistics. Release 5. Chicago: SPSS, 1992.
3
Reynolds T . Breast cancer prognostic factors -- the search goes on. J Natl Cancer Inst 1994;86:480-483
CrossRef | Web of Science | Medline4
Ewertz M ,Gillanders S ,Meyer L ,Zedeler K . Survival of breast cancer patients in relation to factors which affect the risk of developing breast cancer. Int J Cancer 1991;49:526-530
CrossRef | Web of Science | Medline
To the Editor:
Muss et al. (May 5 issue)1 report that overexpression of c-erbB-2 (HER-2/neu) identifies those patients with node-positive breast cancer who will derive the most benefit from higher doses of cyclophosphamide, doxorubicin, and fluorouracil. They offer the caveat that estimation of the overexpression of this gene may not be ideal for the purposes discussed, since “no standardized assay is available.”
We call your attention to our recent article describing a standardized Southern blot-based assay in which DNA derived from human breast-cancer cells and hybridized simultaneously to HER-2/neu and to a single-copy, reference-gene probe accurately measures the gene-copy numbers in tumors2. We have shown that copy numbers of the HER-2/neu gene in patients can be readily obtained and compared with standards derived from breast-cancer cell-line DNA with established, amplified copy numbers. As stated by Muss et al.1 and also shown elsewhere, immunohistochemical analysis of c-erbB-2 expression correlates closely with HER-2/neu gene amplification3-5.
A large retrospective study using simultaneous hybridization for quantitation may accurately stratify groups of patients according to whether they will respond or will not respond to chemotherapy1. The quantitative nature of assays of gene-copy numbers, as compared with qualitative immunohistochemical staining, should delineate even fine differences among groups of patients and thus offer an opportunity for standardization.
5 ReferencesDaniel H. Farkas, Ph.D.
William Beaumont Hospital, Royal Oak, MI 48073-6769Robert M. Umek, Ph.D.
University of Virginia, Charlottesville, VA 229031
Muss HB ,Thor AD ,Berry DA , et al. c-erbB-2 Expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994;330:1260-1266
Full Text | Web of Science | Medline2
Farkas DH ,McMahon LW ,Cai P ,Umek RM . Improved quantitation of HER-2/neu gene copy number in breast tumor-derived DNA samples. Am J Clin Pathol 1993;100:444-450
Web of Science | Medline3
Liu E ,Thor A ,He M ,Barcos M ,Ljung BM ,Benz C . The HER2 (c-erbB-2) oncogene is frequently amplified in in situ carcinomas of the breast. Oncogene 1992;7:1027-1032
Web of Science | Medline4
Kerns BJ ,Pence JC ,Huper G ,Kinney RB ,Iglehart JD . c-erbB-2 Expression in breast cancer detected by immunoblotting and immunohistochemistry. J Histochem Cytochem 1990;38:1823-1830
CrossRef | Web of Science | Medline5
Naber SP ,Tsutsumi Y ,Yin S , et al. Strategies for the analysis of oncogene overexpression: studies of the neu oncogene in breast carcinoma. Am J Clin Pathol 1990;94:125-136
Web of Science | Medline
To the Editor:
With regard to the article by Muss et al.1 on the expression of c-erbB-2 and the response to adjuvant chemotherapy in patients with breast cancer, this report concerns a subgroup of patients from the larger Cancer and Leukemia Group B trial of dose intensity in adjuvant breast cancer. It seems that the subgroup chosen for the investigation of c-erbB-2 expression is somehow not representative of the total population from which the subgroup was culled, thus calling into question the validity of the results. That is, in the multivariate Cox model discussed on page 1263 and presented in Table 4, the dose of cyclophosphamide, doxorubicin, and fluorouracil (CAF) is not predictive of overall or disease-free survival. The variable prospectively studied and found to be a determinant of survival in the parent study2 does not hold up in a Cox regression analysis of the subgroup of patients. I wonder if the authors would comment on this discrepancy.
A potentially more important point relates to the implications of the study by Muss et al. if the results turn out to be reproducible. Although adjuvant chemotherapy for stage II breast cancer is clearly efficacious, one must admit that the benefit is small. This small benefit may indicate either that all women benefited a little bit or that a small subgroup of women benefited a lot. The study of Muss et al. obviously did not include a no-treatment group, but if there is a threshold effect for chemotherapy,2 then the low-dose group may be the same as a no-treatment group. One therefore wonders whether, for the 70 percent of patients with low c-erbB-2 expression, adjuvant chemotherapy is doing nothing, whereas for the patients with c-erbB-2 overexpression, chemotherapy has a large benefit. Addressing this possibility would be ethically problematic, but it would be important to know whether the majority of patients with stage II disease do not benefit from chemotherapy and hence may safely be spared its toxicity.
2 ReferencesBriggs W. Morrison, M.D.
Dana-Farber Cancer Institute, Boston, MA 021151
Muss HB ,Thor AD ,Berry DA , et al. c-erbB-2 Expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994;330:1260-1266
Full Text | Web of Science | Medline2
Wood WC ,Budman DR ,Korzun AH , et al. Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N Engl J Med 1994;330:1253-1259
Full Text | Web of Science | Medline
To the Editor:
In discussing the study by Muss et al.,1 Goldhirsch and Gelber incorrectly state in their editorial (May 5 issue)2 that the “c-erbB-2 results represent an unanticipated secondary outcome.” Muss et al. note in their introduction that their objective was to analyze indirect measures of proliferative activity, including c-erbB-2 expression. Goldhirsch and Gelber were apparently misled by the authors' use of expected effects of DNA content to determine an appropriate sample size.
Goldhirsch and Gelber are also incorrect in stating that the c-erbB-2 data were transformed to enhance their predictive value; transformations were used only for the number of positive nodes, estrogen-receptor content, and S-phase fraction (Table 3 of Muss et al.). The concern about the cutoff used is also unfounded, since the data were analyzed as for a continuous variable; the cutoff was used only to “portray these data graphically.”1
I take issue with two more statements made by Goldhirsch and Gelber. First, in discussing the paper by Wood et al.,3 they note a trend favoring the high-intensity regimen. However, the overall difference in survival at three years was only 2 percent (95 percent confidence interval, -2 to 6 percent), as calculated from the individual standard errors listed in Table 34. Second, they cite a study by Buzzoni et al.5 as evidence that doxorubicin alone or in combination is the most effective adjuvant therapy for some patients. However, those results may also be interpreted to support the data of Wood et al. indicating that adjuvant therapy requires a threshold level of dose intensity (which seems more likely than a dose-response effect), since the alternating-therapy group in the study by Buzzoni et al. used very low dose intensities of CMF and doxorubicin if one considers each regimen separately.
5 ReferencesCarl D. Atkins, M.D.
242 Merrick Rd., Rockville Centre, NY 115701
Muss HB ,Thor AD ,Berry DA , et al. c-erbB-2 Expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994;330:1260-1266
Full Text | Web of Science | Medline2
Goldhirsch A ,Gelber RD . Understanding adjuvant chemotherapy for breast cancer. N Engl J Med 1994;330:1308-1309
Full Text | Web of Science | Medline3
Wood WC ,Budman DR ,Korzun AH , et al. Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N Engl J Med 1994;330:1253-1259
Full Text | Web of Science | Medline4
Longitudinal studies and the use of the life table. In: Colton T. Statistics in medicine. Boston: Little, Brown, 1974:249.
5
Buzzoni R ,Bonadonna G ,Valagussa P ,Zambetti M . Adjuvant chemotherapy with doxorubicin plus cyclophosphamide, methotrexate, and fluorouracil in the treatment of resectable breast cancer with more than three positive axillary nodes. J Clin Oncol 1991;9:2134-2140
Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: Drs. Coppin and Goldie call attention to the finding of a significant difference in outcome for the patients in group 1 (high dose intensity) or group 2 (moderate dose intensity) as compared with the patients in group 3 (low dose intensity). Group 3 received half the cumulative dose of the first two groups. The difference in outcome between groups 1 and 2 was not significant. Their suggestion that the total dose is the underlying basis for the observed differences would be in agreement with one of the a priori design hypotheses of this clinical trial. However, the difference between groups 1 and 2 (albeit not significant) and the incremental benefits of higher dose intensity in the subgroups with overexpression of c-erbB-21 allow the consideration of a nonlinear dose-response effect as another explanation.
Drs. Muller, Gleiter, and Gundert-Remy suggest that the collection of randomly timed plasma samples from each patient would allow pharmacokinetic-pharmacodynamic analyses in a chemotherapy trial. Variations in individual plasma levels and clearance rates should be balanced across the groups in a randomized trial so that the question addressed would not be influenced by the data obtained and the answer not be biased by the lack of individual plasma levels. The issue of tailored therapy based on individual clearance rates and the calculated area under the concentration-time curve was not addressed in this trial. This fruitful area of research is addressed in Cancer and Leukemia Group B protocol C8944, a trial involving patients with stage III breast cancer that was designed to allow the correlation of pharmacokinetic data with the pathological and clinical measurement of response.
Drs. Melnychuk and Panasci are concerned about the adequacy of a median follow-up of 3.4 years. They illustrate their concern by citing the trial of adjuvant CMF therapy by Bonadonna et al.2: “In their first study, which had a median follow-up of 27 months, postmenopausal women with stage II breast cancer treated with CMF had a survival advantage as compared with controls.” They are mistaken. The report Melnychuk and Panasci cite had a follow-up of less than 14 months. The advantage seen was only in the disease-free period. No survival advantage was observed at the time of the first or subsequent reports, as Dr. Bonadonna was careful to state2. Although further follow-up is an important aspect of our clinical trial, the survival differences at a median follow-up of 40 months are not likely to be reserved. We continue to monitor outcomes in this trial.
Dr. Sauter believes the risk of cardiac toxicity from doxorubicin is unwarranted in women with fewer than four axillary lymph nodes showing regional metastases. An intergroup study (Southwest Oncology Group, Eastern Cooperative Oncology Group, and Cancer and Leukemia Group B) recently concluded the accrual of more than 4000 women with breast cancer and negative nodes for a comparison of CMF with CAF. If there is an unanticipated late detrimental effect, it will come to light in the ongoing follow-up of this trial. Women with uninvolved axillary lymph nodes or only one to three positive nodes have a lower rate of failure, so determination of the toxicity-benefit rate from the use of doxorubicin in these patient groups will require longer follow-up for adequate assessment. Data supporting clinically important myocardial damage in otherwise healthy persons treated with these doses of doxorubicin are lacking. Dr. Sauter's note of caution is precisely the reason for randomized trials with large numbers of patients -- to dissect out the long-term benefits and risks of such changes as the introduction of anthracyclines into adjuvant therapy for breast cancer as factual rather than hypothetical risk balances.
Drs. Garey and Lehrer note an association between family history and a reduced time to recurrence in 137 women receiving radiation therapy. Such an association must be balanced against other series, such as Anderson and Badzioch's comparison of 556 patients with familial breast cancer and 4551 patients with breast cancer, in which there was no association between family history and outcome when the stage of disease was controlled for3. At the annual meeting of the American Society of Clinical Oncology, Dr. Henry Lynch and colleagues reported that breast cancers in patients with BRCA-1 linkages have “adverse pathology indicators but, paradoxically, better survival.”4
4 ReferencesWilliam C. Wood, M.D.
Emory University School of Medicine, Atlanta, GA 30322Daniel Budman, M.D.
North Shore University Hospital, Manhasset, NY 11030I. Craig Henderson, M.D.
University of California, San Francisco, School of Medicine, San Francisco, CA 941431
Muss HB ,Thor AD ,Berry DA , et al. c-erbB-2 Expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994;330:1260-1266
Full Text | Web of Science | Medline2
Bonadonna G ,Brusamolino E ,Valagussa P , et al. Combination chemotherapy as an adjuvant treatment in operable breast cancer. N Engl J Med 1976;294:405-410
Full Text | Web of Science | Medline3
Anderson DE ,Badzioch MD . Survival in familial breast cancer patients. Cancer 1986;58:360-365
CrossRef | Web of Science | Medline4
Lynch HT ,Marcus J ,Watson P ,Page D . Distinctive clinicopathologic features of BRCAl-linked hereditary breast cancer. Prog Proc Am Soc Clin Oncol 1994;13:56-56 abstract.
Author/Editor Response
We appreciate the interest in our article on the relation between treatment outcome and c-erbB-2 expression in women with node-positive breast cancer1. We recognize that there are several methods of identifying alterations in c-erbB-2 gene-copy numbers, including Southern blotting, Northern blotting, fluorescence in situ hybridization, and differential polymerase-chain-reaction techniques. The Southern blot assay described by Drs. Farkas and Umek2 appears most promising. Nevertheless, we still believe there is no widely accepted, standardized technique for c-erbB-2 measurement.
Dr. Morrison points out that the dose of CAF was not a significant predictor of overall disease-free survival in our sample1. This observation is an essential aspect of our conclusion: the dose of CAF matters only through its interaction with c-erbB-2 expression. Our sample was representative of the larger trial3 (see Table 1 of our article). Accounting for c-erbB-2 in our sample shows that the benefit of increasing dose intensity is restricted to the subpopulation with c-erbB-2 overexpression. In this sense, accounting for c-erbB-2 makes clearer the part played by the dose of CAF.
Dr. Morrison's more important point is the essence of our paper: that the benefit of increasing the dose of CAF is restricted to patients with c-erbB-2 overexpression. As for whether low-dose CAF is the same as no dose, we do not know. Evidence from Tsai et al.4 suggests that the concentration of doxorubicin required to inhibit lung-cancer cell lines by 50 percent increases with increasing c-erbB-2 expression. This suggests that low-dose doxorubicin may be adequate for “underexpressers,” but that increasing dose or intensity may not result in additional benefit. Only clinical trials comparing a low-dose group with a no-treatment group could resolve this question satisfactorily.
Dr. Atkins is correct in stating that the size of our sample was based on predicted differences in outcome in relation to DNA content and that we analyzed c-erbB-2 expression as a continuous variable, using a cutoff only to portray the data graphically; however, the relation of c-erbB-2 expression to outcome was unanticipated.
As we stated in our article, we believe that confirmation of this work is necessary. Both the Cancer and Leukemia Group B and other cooperative groups are investigating further the relation of c-erbB-2 expression to treatment outcome in patients treated with different regimens of adjuvant combination chemotherapy, as well as adjuvant endocrine therapy.
4 ReferencesHyman B. Muss, M.D.
Bowman Gray School of Medicine, Winston-Salem, NC 27157Ann D. Thor, M.D.
University of Vermont School of Medicine, Burlington, VT 05405Donald A. Berry, Ph.D.
Duke University Medical Center, Durham, NC 277081
Muss HB ,Thor AD ,Berry DA , et al. c-erbB-2 Expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994;330:1260-1266
Full Text | Web of Science | Medline2
Farkas DH ,McMahon LW ,Cai P ,Umek RM . Improved quantitation of HER-2/neu gene copy number in breast tumor-derived DNA samples. Am J Clin Pathol 1993;100:444-450
Web of Science | Medline3
Wood WC ,Budman DR ,Korzun AH , et al. Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N Engl J Med 1994;330:1253-1259
Full Text | Web of Science | Medline4
Tsai C-M ,Chang K-T ,Perng R-P , et al. Correlation of intrinsic chemoresistance of non-small-cell lung cancer cell lines with Her-2/neu gene expression but not with ras gene mutations. J Natl Cancer Inst 1993;85:897-901
CrossRef | Web of Science | Medline
Author/Editor Response
Dr. Atkins failed to understand some of the points in our editorial on the Cancer and Leukemia Group B investigations1,2. The results of the clinical trial reported by Wood et al.1 are relevant to patient care today for the reasons we stated. In contrast, the results of the substudy presented by Muss et al.2 must be viewed as the basis for future investigations and are not ready for prime time for a variety of reasons. We stand by our statement that “c-erbB-2 results represent an unanticipated secondary outcome” to remind readers that the c-erbB-2 analysis yielded the most statistically significant and provocative results only among the four “new molecular prognostic factors” tested (DNA content, S-phase fraction, c-erbB-2 expression, and p53). Because a picture is worth a thousand words, we continue to be cautious about the clinical (and biologic) relevance of a retrospectively defined cutoff used to “portray these data graphically.” We wonder whether the selection of other cutoff values along the continuum would have produced figures with comparable visual impact. Readers should not be misled into thinking that c-erbB-2 expression is the key measure defining responsiveness to treatment; it is only the one that produced the most impressive results in this investigation.
We used the trial by Buzzoni et al.3 in our editorial as an example of the most effective use of anthracyclines when combined with the CMF regimen. The trial does not provide a useful example for extrapolating dose-response relations, because the drugs used in each treatment course in the two treatment groups were given at an identical dose intensity. We take the point made by Dr. Atkins that if one considers doxorubicin to be superior to CMF in terms of cells killed, then the issue of dose intensity becomes important. However, there are no data to support the assumption that the doses of anthracyclines currently used are more effective than full-dose CMF4.
4 ReferencesAron Goldhirsch, M.D.
International Breast Cancer Study Group, 6900 Lugano, SwitzerlandRichard D. Gelber, Ph.D.
Dana-Farber Cancer Institute, Boston, MA 021151
Wood WC ,Budman DR ,Korzun AH , et al. Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N Engl J Med 1994;330:1253-1259
Full Text | Web of Science | Medline2
Muss HB ,Thor AD ,Berry DA , et al. c-erbB-2 Expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 1994;330:1260-1266
Full Text | Web of Science | Medline3
Buzzoni R ,Bonadonna G ,Valagussa P ,Zambetti M . Adjuvant chemotherapy with doxorubicin plus cyclophosphamide, methotrexate, and fluorouracil in the treatment of resectable breast cancer with more than three positive axillary nodes. J Clin Oncol 1991;9:2134-2140
Web of Science | Medline4
Shapiro CL ,Henderson IC . Late cardiac effects of adjuvant therapy: too soon to tell? Ann Oncol 1994;5:196-198
Web of Science | Medline
