This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Sleeper, A. M. Articles by Thomé, S. D. Search for Related Content PubMed PubMed Citation Articles by Sleeper, A. M. Articles by Thomé, S. D. Social Bookmarking                            What's this?

Assessing Adjuvant Breast Cancer Therapy Benefit

Ravenel Oncology Center, Martinsville, VA

To the Editor:The article by Loprinzi and Thomé¹ on adjuvant systemic therapy for primary breast cancer offers minimal assistance in deciding whether to give adjuvant chemotherapy in breast cancer. There is a general consensus that patients with node-positive disease (especially estrogen receptor–negative) and large tumors benefit from chemotherapy. The major problem arises in dealing with the increasingly frequent mammogram-discovered, small, node-negative lesion. The authors’ approach is to take the collected data from the Early Breast Cancer Trialists articles^2,3 (which stratify patients by presence or absence of nodes but not by tumor size) and assume that the annual proportional reduction in risk applies equally to all tumors of a given nodal and estrogen receptor status, irrespective of size. This information is then fed into an opinion-generated table of baseline risk as a function of tumor size and nodal status to generate tables of clinical benefit (Tables 7 and 8).¹

There are a number of problems with this approach: (1) The input baseline risk (Table 2) is only opinion generated. Correlation is claimed with Dr Peter Radvin’s "Aduvant!" program based on Surveillance, Epidemiology, and End-Results data and a proprietary formula (Fig 1). However, close inspection of that figure in the low-risk, high-survival range shows significant differences of magnitudes similar to the claimed benefits of chemotherapy. (2) The assumption that tumors of different sizes have the same annual proportional reduction in risk is questionable. The data analyzed by the Early Breast Cancer Trialists are not stratified by tumor size. Furthermore, the data are based on studies a decade old that include mostly patients with larger tumors or node-positive tumors. Extrapolation to small node-negative tumors could well be in error. Perhaps there are threshhold effects below which metastasis is exceedingly unlikely. (3) Table 8 is filled with rows of entries for several different ranges of involved nodes. These data are not particularly useful because all of these patients are treated anyway. The issue is the small node-negative tumors where the data are most suspect. (4) The patients in question with small node-negative tumors have low recurrence rates. Other variables not included in this model, such as histology, could have large relative effects. (5) The information on paclitaxel, based upon a single abstract, has now become highly suspect.⁴

As a medical oncologist I am still left with the problem of whether to recommend treatment of women with small, node-negative tumors at considerable toxicities and with minimal statistical benefit. Certainly the marginal statistical benefit implies that we are subjecting many more women to toxicity than will benefit from therapy. The answer will come from better prognostic markers, such as Braun et al’s⁵ work with bone marrow markers or perhaps ultimately from gene expression arrays.⁶ More effort needs to be directed in these areas rather than continuing to rehash old data.

REFERENCES

1. Loprinzi CL, Thomé SD: Understanding the utility of adjuvant systemic therapy for primary breast cancer. J Clin Oncol 19: 972-979, 2001[Abstract/Free Full Text]

2. Early Breast Cancer Trialists’ Collaborative Group: Polychemotherapy for early breast cancer: An overview of the randomised trials. Lancet 352:930-942, 1998

3. Early Breast Cancer Trialists’ Collaborative Group: Tamoxifen for early breast cancer: An overview of the randomised trials. Lancet 351:1451-1467, 1998

4. National Institutes of Health: Adjuvantive therapy for breast cancer. NIH Consensus Statement Online 2000, November 1-3, 17:1-23, 2000

5. Braun S, Cevatli BS, Assemi C, et al: Comparative analysis of micrometastasis to the bone marrow and lymph nodes of node-negative breast cancer patients receiving no adjuvant therapy. J Clin Oncol 19: 1468-1475, 2001[Abstract/Free Full Text]

6. Hedenfalk I, Duggan D, Chen Y, et al: Gene-expression profiles in hereditary breast cencer. N Engl J Med 348: 539-548, 2001

Response

Gundersen Clinic, Ltd, La Crosse, WI

To the Editor:In the article by Loprinzi and Thomé,¹ the authors describe, as an example, a 55-year-old, node-negative, estrogen receptor–positive breast cancer patient who will improve her 10-year disease-free survival from 70% to 79% with combined adjuvant therapy while her chance of being free of disease recurrence will actually improve to 87%. In other words, she apparently benefits an additional 8% because she will die of another cause. Although at first glance these distinctions are confusing, such calculations underscore the obvious importance of considering age and comorbidity in advising patients and also in realizing that although the Numeracy program, according to the authors, "can factor in the mortality that would be associated with a woman who was otherwise healthy, based on causes of mortality other than breast cancer," the online program currently available (http://mhs.mayo.edu/adjuvant) does not.

REFERENCES

1. Loprinzi CL, Thomé SD: Understanding the utility of adjuvant systemic therapy of primary breast cancer. J Clin Oncol 19: 972-979, 2001

Response

Mayo Clinic, Rochester, MN

In Reply:We appreciate the opportunity to respond to the letters authored by Dr Arthur M. Sleeper and Drs Steven M. Sorscher and Leah L. Dietrich. First, Sorscher and Dietrich insightfully raise a seemingly confounding issue dealing with why the overview analysis reports an improved outcome (in terms of annual proportional reductions in risk recurrence) for disease-free survivals versus overall survivals.¹ This reported differential effect suggests, in the patient example cited, that combined adjuvant therapy provides an 8% additional benefit for survival (as opposed to disease-free survival). There are at least three reasons for this discrepancy, one of which, as Sorscher and Dietrich surmise, is that some patients, despite developing recurrent breast cancer, will die of causes other than breast cancer. Two other reasons, however, are that (1) the overview analysis counts a contralateral breast cancer as a recurrence and (2) the delay between recurrence and death makes the benefits sound more robust regarding recurrence, as opposed to death. As might be predicted, the latter reason should become less prominent as the overview data further mature. This effect is evident in the most recent overview analyses, which were recently presented at a recent National Institutes of Health Consensus Development Conference but which have yet to be published.

It is noteworthy, as we discussed in our publication, that we use the terms 10-year survival and 10-year disease-free survival interchangeably when talking about eventual prognosis, given that these two percentages are relatively close to each other. It is true that the online version of Numeracy, at http://mhs.mayo.edu/adjuvant, does not allow for factoring in comorbid conditions. For this online version of Numeracy, we felt that a simple pull-down menu with automated data entry would be best for most expected users. Nonetheless, there is an available Numeracy spreadsheet program that does have the ability to change 10-year baseline risk estimates and efficacy estimates at will. We are happy to provide interested physicians with a copy of this spreadsheet program (please request at email: cloprinzi@mayo.edu).

Moving on to respond to comments from Dr Sleeper, let us address the five problems that he lists. Sleeper first takes issue with the baseline risk that we used. We admit that this risk is opinion generated. However, it is not individual physician opinion-generated information, which, as illustrated in Table 2 of our article² and in a previous publication,³ is very diverse. Rather, it is an average opinion from a number of clinicians who regularly see patients with breast cancer. We are impressed with the remarkable correlation between prognosis seen in our program with that seen with the method from Ravdin et al.⁴ With regard to the relatively modest differences in opinions between our baseline estimates and those of Ravdin et al,⁴ it is not clear which one is more or less correct. Although we are hopeful that more precise baseline prognoses will be generated in the future, we are unaware of more helpful baseline prognosis information to use for individual patients.

Sleeper’s next concern deals with whether annual proportional risk reduction is an accurate way of understanding chemotherapy effects across patients with different tumor sizes and different nodal situations. We note that we did not define this relative or proportional mechanism of describing the benefits of adjuvant therapy, a mechanism that has been routinely used to describe adjuvant treatment benefits in recent reports of clinical trials and in adjuvant meta-analyses and overviews. Rather, our article sought to better translate individual patient proportional benefits into absolute benefits.

Sleeper’s next concern deals with Table 8. In contrast to Sleeper’s assertion, this table contains information regarding both patients with and without involved axillary lymph nodes. We agree with Sleeper that the information regarding small, node-negative cancers might be most helpful for allowing a patient and her physician to better understand the potential benefits from systemic therapy. In addition, however, we believe that understanding the relative benefits of adjuvant therapy in node-positive patients would be helpful for many patients so that they can better know the magnitude of benefit for which they are receiving toxic therapy.

The fourth problem that Sleeper notes is that other prognostic variables, such as histology, were not included in our baseline prognostic model. We will admit that there are many different prognostic tools that have been studied in groups of patients. Nonetheless, given the vast diversity of opinions regarding the best two established features (that being the number of involved axillary lymph nodes and tumor size), the use of less robust prognostic factors, in individual patient cases, is not established. Sleeper is referred to Fig 4 in a previously published article,² which illustrates the limited value from adding weaker prognostic factor information to prognostic opinions based on axillary lymph node status and tumor size.

Sleeper’s last concern deals with the fact that paclitaxel is not fully established, effective therapy for patients in the adjuvant setting. We agree, as indicated by reminding him that this was noted in our article on four separate occasions.

Sleeper ends his letter looking to the future when, hopefully, better prognostic information will be available, as opposed to trying to better understand how to use the data that are accessible to us today. We, too, look forward to the future when better prognostic information should be available. In contrast to Sleeper, however, we do believe that there is room for a better understanding of the prognostic information that is currently available so as to better be able to inform and treat those patients we see in our practice today. The multiple, positive, unsolicited comments that we have received from colleagues since the publication of this article indicates to us that compiling available information in an accessible manner is worthwhile. Sometimes old hash does taste better after rewarming.

REFERENCES

1. Early Breast Cancer Trialists’ Collaborative Group: Polychemotherapy for early breast cancer: An overview of the randomised trials. Lancet 352:930-942, 1998

2. Loprinzi CL, Thomé SD: Understanding the utility of adjuvant systemic therapy for primary breast cancer. J Clin Oncol 19: 972-979, 2001

3. Loprinzi CL, Ravdin PM, De Laurentiis M, et al: Do American oncologists know how to use prognostic variables for patients with newly diagnosed primary breast cancer? J Clin Oncol 12: 1422-1426, 1994[Abstract]

4. Ravdin PM, Siminoff LA, Davis GJ, et al: Computer program to assist in making decisions about adjuvant therapy for women with early breast cancer. J Clin Oncol 19: 980-991, 2001[Abstract/Free Full Text]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Sleeper, A. M. Articles by Thomé, S. D. Search for Related Content PubMed PubMed Citation Articles by Sleeper, A. M. Articles by Thomé, S. D. Social Bookmarking                            What's this?