Over 200,000 women in the US will be diagnosed with ]]>breast cancer]]> in 2005. After lung cancer, breast cancer is the second leading cause of cancer death among American women.

The American Cancer Society recommends that all women aged 40 and older have a yearly ]]>mammogram]]> (x-ray of the breasts) and clinical breast exam (examination of the breasts by a health professional), and all women in their 20s and 30s have a clinical breast exam at least every three years.

Women with breast cancers detected by screening mammography have better survival rates than those detected symptomatically, but there is some debate as to why. Many researchers believe the favorable outcomes associated with screening mammography are entirely due to “lead-time bias,” a study bias that favors mammography because of lead time, or the time between when the tumor was detected by mammography and when the tumor would have been detected symptomatically.

But a new study in the August 17, 2005 issue of the Journal of the National Cancer Institute adjusted for breast cancer stage and other tumor characteristics in order to account for lead time. The researchers found that, even after adjusting for lead time, women whose breast cancers were detected through screening had a survival advantage over women whose breast cancers were detected symptomatically.

About the Study

This study included data from two large breast cancer screening trials, the Health Insurance Plan (HIP) of New York, which began in the 1960s, and the Canadian National Breast Cancer Screening Studies (CNBSS), which began in 1980.

The HIP trial included approximately 62,000 women aged 40-64 who were randomly assigned to receive four yearly clinical breast exams and mammograms (screening group) or “usual practice” (control group). The CNBSS included approximately 90,000 women aged 40-59 were randomly assigned to receive 4-5 annual clinical breast exams and mammograms (screening group) or one clinical breast exam but no mammogram (control group).

In women who developed breast cancer, researchers collected data on the tumor size (small or large), cancer stage (I, II, or III/IV, with III/IV being the most advanced), and whether the cancer had spread to lymph nodes. The researchers classified the cancers as screening-detected (those detected by a screening examination), interval/incident (those diagnosed after a negative screening examination), or symptom-detected (those detected in the control group).

The researchers kept track of time of diagnosis to time of death in the women with breast cancer.

The Findings

After a five-year follow-up, 608 women in the HIP trial and 1,319 women in the CNBSS trial were diagnosed with breast cancer.

As expected, the screening-detected cancers were detected at an earlier stage than other cancers. For example, in the HIP trial 76% of the screening-detected cancers were stage I (the least advanced), whereas 51% of the interval/incident cancers and 43% of the symptom-detected cancers were in stage I. In addition, the screening-detected cancers tended to be smaller and were less likely to have spread to the lymph nodes, compared with other cancers.

Not surprisingly, the women whose cancers were screening-detected had the longest survival overall. But interestingly, even after comparing survival rates in women with tumors at the same stage, size, and lymph node status, the women with screening-detected cancers still had the longest survival, though by a smaller margin.

After adjusting for tumor stage (controlling for lead-time bias), women with incident/interval and symptom-detected cancers had a 53% and 36% greater risk of death than women with screening-detected cancers.

How Does This Affect You?

These results indicate that how breast cancer is detected can affect the outcome of the disease. In other words, this study suggests that if one woman’s breast cancer was diagnosed by a mammogram and another’s was diagnosed after symptoms appeared, the woman who had the mammogram would be more likely to survive longer, even if their cancers were seemingly identical.

Does this mean that some aspect of breast cancer screening itself benefits women diagnosed with breast cancer? That is unlikely. A more plausible explanation is that screening detects cancers with some other unknown characteristic, for instance the tendency to be slower growing, making them less deadly.

If you or someone you know is diagnosed with breast cancer, the doctor may consider how the breast cancer was detected when determining the likely outcome of the cancer. All else equal, women whose breast cancer was diagnosed with screening may be better off than those diagnosed after symptoms appeared.

Also, this study may have an impact on clinical research. The study’s authors suggest that other clinical trials that are looking at breast cancer treatments should record and adjust for how the cancer was diagnosed, since it appears that method of detection may affect outcomes.