| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Commentary |
1 Division of Breast Imaging, University of California–San Francisco, 1600 Divisadero St., Ste. H2801, San Francisco, CA 94115.
Received October 18, 2004; accepted after revision October 19, 2004.
Address correspondence to J. W. T. Leung
(Jessica.Leung{at}ucsfmedctr.org).
OBJECTIVE. This commentary discusses the retrospective cohort study by Barth et al. studying the effect of screening mammography on morbidity of breast cancer treatment.
CONCLUSION. Barth et al. found that cancers identified at screening mammography (versus those detected by physical examination) were smaller and more frequently node-negative. Hence, the former group of breast cancer patients was more likely to be treated with less morbid therapy. This evidence-based study contributes to the clinical practice of screening mammography.
There has been much debate regarding the value of screening mammography. Randomized controlled trials have shown that screening mammography reduces the mortality rate of breast cancer [1, 2]. Nevertheless, debate continues regarding the true clinical benefit conferred by screening mammography, particularly for women at the extremes of age [2]. Aside from the absolute mortality of breast cancer, breast cancer treatment is associated with significant morbidity, both physical and emotional, and financial costs.
The goal of screening mammography is the detection of breast cancer in the preclinical stage. On average, when compared with physical examination, screening mammography detects smaller cancers that are less likely to have metastasized to the axillary nodes. Generally speaking, the smaller the cancer, and the more likely it is to be node-negative, the more likely that the patient will be treated with less morbid and less costly therapies. Therefore, it follows that screen-detected breast cancers can be treated more often with breast conservation surgery rather than mastectomy and no chemotherapy. Relatively little research concerning the influence of the mode of detection of breast cancer on subsequent treatment has been published.
Barth et al. [3] investigated whether breast cancer detected by screening mammography is associated with less toxic therapy compared with breast cancer detected by physical examination. To achieve this end, these researchers conducted a retrospective cohort observational study consisting of 992 women with invasive breast cancer diagnosed at a single institution between 1990 and 2001. The study cohort was approximately equally divided between patients with breast cancer detected by screening mammography (n = 460) and patients with breast cancer detected by physical examination (n = 532).
The primary findings were that patients whose cancers were detected by screening mammography were more likely to be treated with breast conservation surgery (56% vs 32%) and less likely to receive chemotherapy (28% vs 56%) than patients whose cancers were detected by physical examination. Presumably, this is because patients in the former group were more frequently node-negative (84% vs 58%) and had smaller tumors (1.5 vs 2.9 cm) than patients in the latter group. The number of cases was relatively large, and the differences between the two groups were sufficiently great. Hence, statistical significance was readily achieved.
The authors stratified their analysis by patient age groups. They recognized that the mortality reduction conferred by screening mammography for women in their fifth decade remains hotly debated [2]. The incidence of breast cancer is lower in this age group, but these cancers tend to be more aggressive. Therefore, it is controversial whether screening mammography actually saves lives in women of this age group. The authors also appropriately noted that data on the efficacy of screening mammography in women 70 years old or greater are very scarce. Hence, they specifically and meaningfully analyzed their results with respect to women at these two age extremes.
In the analysis of stratified data, there were 258 women aged 40–49 and 230 women aged 70 or greater. When the tumor was detected by screening mammography, the mean tumor size was smaller and the percentage of node-negative tumors was higher for both age groups than tumors detected by physical examination. Specifically, when comparing tumors detected by screening mammography versus those detected by physical examination in women aged 40–49, the mean tumor size was 1.6 versus 2.8 cm, and the percentage of node-negative tumors was 74% versus 52%. Similarly, the numbers for women 70 or older were 1.4 versus 2.9 cm and 88% versus 72%, respectively.
Consequently, the number of patients with screen-detected cancers who were treated with breast conservation surgery was higher, and the number of patients with screen-detected cancers receiving chemotherapy was lower. In the younger age group, the percentages of patients treated with breast conservation surgery for tumors detected by screening mammography versus physical examination were 52% versus 38%. In the older age group, the percentages were 66% versus 27%, respectively. In the younger age group, the percentage of patients receiving chemotherapy was 57% for those with cancers detected by screening mammography versus 75% for those with cancers detected by physical examination. In the older age group, the numbers were 4% versus 15%, respectively. Presumably, the overall lower rates of chemotherapy administration in older women were related to comorbid conditions and reduced life expectancy compared with the life expectancy of the younger women.
The authors used a relational database and performed multivariate analysis to adjust for potential confounders. They adjusted for various clinical, demographic, and pathologic factors, including functional health status, as measured by Karnofsky performance scale scores. The patient's baseline functional status affects the type of treatment offered and tolerated, so consideration of this factor as a potential confounding variable was appropriate.
After adjustment for patient covariates, including age, menopausal status, and functional health status, women with physical examination–detected tumors were 2.9 times more likely to be treated with chemotherapy and 2.5 times more likely to be treated with mastectomy. These findings held true statistically when analysis was stratified for age group 70 or greater. In age group 40–49, women with cancers detected by physical examination were more likely to receive chemotherapy, but this observation did not satisfy tests of statistical significance when comparing whether these women were more likely to undergo mastectomy.
The design of this study was crisp, tailored to answer a specific research question. The results were solid, and the analysis was statistically sound. Modeling (logistic regression) and stratification were used appropriately to adjust for potential confounding variables. The findings were thoughtfully discussed in the context of previously published studies. The unique strengths of this study were as follows: a specific and well-defined research question; large study cohort, achieving statistical significance; and multivariate analysis of potential confounders, thereby showing that method of detection of breast cancer was an independent predictor variable of subsequent type of cancer treatment.
As is the case with other single-institution reports, the ability to generalize on the basis of the study findings may be limited. To address this potential limitation, the authors compared the characteristics of the study cohort to those of the cancer patients in the New Hampshire Mammography Network, a population-based mammography registry that captured over 90% of women undergoing mammography in the state [4, 5]. They found that the study cohort and the cancer patients in the registry were very similar, proving that the results were at least applicable to the state of New Hampshire. To be strict, the comparison reported only the variables of age and menopausal status and did not present any data on other important variables that affected breast cancer, such as family history, race, or socioeconomic status. Nevertheless, the comparison showed that the findings were likely not limited to one academic center.
This study included only women with invasive breast cancer, and not those with ductal carcinoma in situ (DCIS), as their primary cancer diagnosis. This design was reasonable, because DCIS is generally not treated with chemotherapy and most cases of DCIS are detected by screening mammography and not by physical examination. Furthermore, what constitutes proper treatment of DCIS remains contested, and screening mammography has been criticized for overdiagnosing DCIS [6]. By limiting their study cohort to patients with invasive cancer (and not only DCIS), the authors enhanced the clinical impact of their results and avoided possible confusing interpretations.
It would have been interesting to know the specific clinical symptoms or signs that contributed to the detection of breast cancer. The authors wrote that the most common sign of breast cancer was a palpable mass, but other signs included indentations, nipple retraction, or nipple discharge. Patients who presented with these other symptoms and signs of breast cancer were included in the physical examination group, along with those who presented with a palpable mass. Data regarding the breakdown of these other clinical findings were not available.
Mortality data were outside the scope of this article. In future studies, it would be revealing to follow these patients after treatment for 5 or 10 years to determine whether there were survival differences among the various treatment groups. We could then ascertain that the less morbid treatments given in the screen-detected cancer group were indeed appropriate. Also in a future study, it would be useful to measure quality of life after treatment to determine if less morbid therapy actually translated into better quality of life.
Overall, the article by Barth et al. [3] answered a specific research question of clinical import. The authors concluded that their findings supported the practice of screening mammography, particularly for women at age extremes in whom the mortality reduction of screening mammography has been debated. Screening mammography is complicated, involving not only medical reasoning but also emotional, medicolegal, and cost considerations. Evidence-based research articles in screening mammography such as the one by Barth et al. enhance our everyday clinical practice.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. L. Ellis, D. Gur, A. H. Klym, and J. H. Sumkin Regarding Trends in Recall, Biopsy, and Positive Biopsy Rates for Screening Mammography Radiology, January 1, 2006; 238(1): 375 - 376. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
