Friday, August 21, 2015
An avoidable catastrophe?
This is an observational study only, and it doesn't completely answer the question, but this analysis in JAMA Oncology certainly poses one. Understanding this is a bit complicated, and remember, I'm not a real doctor, I'm a doctor of philosophy, but let me tell you what it means to me.
Narod et al used a registry called the Surveillance, Epidemiology and End Results (SEER) database to see what happened over 10 and 20 years to women who were diagnosed with so-called Ductal Carcinoma in Situ (DCIS). These are clusters of abnormal cells inside the milk duct. Until the widespread use of screening mammography, they were essentially never detected. There's no lump to feel and no symptoms. But since mammography grew more and more sensitive, radiologists started seeing them frequently.
When one is found, the woman is told that she has breast cancer and the alternatives are presented as lumpectomy -- just removing the lesion, but it's a misnomer since there is no lump -- or total mastectomy. Some women have both breasts removed just to be on the safe side. And often they undergo radiation "therapy" as well.
But is this really cancer? And is it really the case, as assumed, that this is how breast cancer begins and these often go on to become invasive and then metastatic? What would happen if they weren't treated and we just waited?
Well, this study can't directly answer the last question because just about every woman with the diagnosis was treated. But it can tell us a few things. The easiest thing to understand is that the choice of treatment -- lumpectomy with or without radiation, or mastectomy -- made no difference in the risk of ultimately dying of breast cancer. The second observation is that Black women were at higher risk than white women of ultimately dying of breast cancer, but there didn't appear to be any difference in how they were treated, so the explanation for this is unknown.
The third observation is that invasive cancer could appear even after mastectomy. While a later occurrence of invasive cancer was associated with higher risk of death, the overall risk for these women was the same as the risk in the general population -- no higher and no lower. The exception is women who were diagnosed with DCIS at a young age. It is also the case that overall, the incidence of invasive cancer in the population has not gone down since doctors started treating DCIS.
What this suggests -- although it doesn't prove it -- is that DCIS is not in fact associated with a higher risk of ultimately getting clinically significant breast cancer, and the best thing to do is just watch and wait. There are other indications from observational data that these lesions often spontaneously remit or just sit there. What this would mean is that tens of thousands of women every year are subjected to fear, disfiguring surgery and frequent complications, at a cost of billions of dollars, for no benefit.
Right now few women are choosing to forego treatment and few doctors are recommending it. But in my view, this does create the so-called equipoise necessary to ethically undertake a randomized controlled trial, if enough women would enroll in it. If the suggestion turns out to be true, we have inflicted major damage on millions of people. Now, surgeons will find it very hard to live with that conclusion I am sure and they will certainly resist it. But it has to be considered.