The Canadian National Breast Screening Study (CNBSS), published last month in the British Medical Journal and widely circulated in US Media, concluded after 25 years that the death rates from breast cancer were the same in women who were screened with mammography and those who were not. They also concluded that 25 percent of the cancers found from screening were overdiagnosed. However, breast imaging specialists who looked at the data found the study to be deeply flawed and disputed its results. National expert Dr. Daniel Kopans, MD, professor of radiology at Harvard Medical School told industry journal Medscape Medical News that he personally reviewed the mammograms in 1990 and found the quality to be poor.
Dr. Catherine Dang, MD associate director the Breast Cancer Risk Reduction Center and a surgeon at the Cedars-Sinai Medical Center in Los Angeles, also told Medscape that “there was no information about tumor state, tumor characteristics or characteristics of study participants. It is unclear whether there was any stratification for age, family history or other risk factors.” The American College of Radiology (ACR), in a statement criticizing several aspects of the study, also noted that the participants in the trial were supposed to be assigned to screened or not screened groups randomly, but in fact had clinical exams before the study and many with suspicious findings were placed in the mammography group. The ACR says that the CNBSS trial “should not be used to create breast cancer screening policy as this would place a great many women at increased risk of dying unnecessarily from breast cancer.”
Specialists in our area also found the study flawed and said it would not affect the way they practice. There was general agreement that early detection of breast cancer is critical to a patient’s treatment and prognosis, and that mammography is still the gold standard, as a starting point, for early detection. Overdiagnosis, the issue raised by the study, cannot be determined because physicians still don’t have the tools to know which cancers are harmless and which are not.
Dr. Lisa Weinstock, a breast imaging specialist and Director of Women’s Digital Imaging of Ridgewood, said early detection of breast cancer is critical for a patient’s outcome. ”Finding cancer early leads to less invasive treatment and fewer complications. It’s not just about survival, it’s about quality of life.” Dr. Weinstock said that women should be screened according to their individual risk profiles, which she prepares for all patients. “Individual characteristics such as family history and breast density determine when and how women should be screened for breast cancer. Breast density can obscure cancers on a mammogram. Who knows how many women in the Canada study had tumors that weren’t seen on the mammogram because they had dense tissue?” Dr. Weinstock said additional imaging such as Ultrasound or MRI can be necessary when mammograms are inconclusive. She initiated a bill in the New Jersey legislature to require women to be notified about the risks of breast density obscuring tumors on mammograms. Beginning later this year, all New Jersey mammography patients will be sent a letter stating that they should ask their doctors about the density results of their mammograms. New Jersey is now the fourteenth state with a breast density notification law. (Disclosure: The author is a consultant for Dr. Weinstock).
Patients are generally referred for mammography screening by internists or gynecologists. Dr. Jennifer Ashton, M.D., F.A.C.O.G., an Englewood based gynecologist and
ABC News Senior Medical Contributor on Good Morning America/ World News with Diane Sawyer and Co-host of The Doctors Show, said the Canada study will not influence her recommendations. “It is well-known that mammography is not the perfect imaging test, but data does support that it can and does detect cancers and saves lives. As a board-certified Ob-Gyn, I will continue to recommend annual screening mammography to my patients 40 and over after an extensive discussion with her regarding the risks, benefits and options to screening.”
Dr. Laura Klein, Medical Director of the Breast Program at The Valley Hospital in Ridgewood, said the conclusion of the Canada study that mammograms lead to over diagnosis may be correct, but the impact cannot be quantified. “I know that some breast cancers will never progress to a life threatening disease, but we don’t know how to select which to treat and which to observe. Until we have those tools, we need to identify and treat all cancers. It is not acceptable to wait until a tumor is large and palpable. We still believe that early diagnosis can lead to cure.”
Twenty percent of cancers found are Ductal Carcinoma in Situ (DCIS) according to Dr. Klein, a condition some experts are saying should be reclassified as not being cancer. Dr. Klein cautions that while that is true, some will progress to invasive disease. “We used to base treatment on the size and lymph node involvement of a tumor,” Dr. Klein said. ”Now we attempt to treat based on individual biology. The more we know about the biology of the tumor, the more we can personalize treatments and use targeted therapies.”
Dr. Klein cautions that reports in the daily papers regarding research studies may present conclusions which are easily misinterpreted. ”I had a patient who read a story about a trial that concluded women with a certain type of cancer didn’t benefit from further treatment such as surgery and radiation. I explained that she didn’t fit the criteria of the study, but she still chose no further treatment.”
Dr. Klein said the Canada Breast Cancer Screening Study will not change her opinion of mammography. ”Mammography continues to be the most accessible, with good sensitivity and specificity, while being a cost effective modality for screening women for breast cancer. The data still supports that mammography saves lives. It is the best screening modality we have.”
Breast cancer is a risk for all women. However, while women in the general US population have about an eight percent chance of developing breast cancer by age 70, the risk is slightly higher among the Ashkenzi Jewish population, largely due to a genetic mutation. According to the National Institutes of Health, it is estimated that in the general US population, between one in 400 and one in 800 people will have a faulty gene putting them at increased risk for breast cancer. In the Ashkenazi Jewish population, the incidence in about one in 40.
By Bracha Schwartz