The Sea of Uncertainty Surrounding DCIS

maize

(pdf file)

http://www.vaoutcomes.org/papers/Welch_Uncertainty.pdf

Beesie

This article was posted by someone else last week.  Here's how I interpret what's being said:

While the authors seem to make a compelling argument, if you read carefully you'll notice that the words are carefully chosen. There's a whole lot of speculation. Some examples:

"Although no one is sure what the probability of progression is, studies of DCIS that were missed at biopsy (1,2) and the autopsy reservoir (3) suggest that the lifetime risk of progression must be considerably less than 50%. "
Translation: We are quoting studies that really can't be interpreted and possibly aren't relevant, and the real truth is that nobody knows what percentage of DCIS progresses to become invasive.  But to make our case, it "must be" less than 50%.  As for the relevance of autopsy studies, who's to say when the DCIS that was found in the autopsy actually developed or how long it's been there?  And how can we know that the DCIS found in the autopsy wouldn't have become invasive within a short period of time, if the individual had not died of other causes?  

. 

"The authors also report that anxiety is associated with exaggerated risk perceptions-many women with DCIS estimate their risk of developing invasive cancer 5 years following treatment to be greater than 50%. Although it's tempting to hope that better educating women (and their doctors) about the actual risks of DCIS would reduce anxiety, it's entirely possible that our measures of risk perception and anxiety are simply markers for the same underlying construct-how women feel about being in the gray zone."
Translation: First, let's notice that the subject has been changed. The discussion is no longer about the percent of DCIS that progresses to IDC. Now the discussion is about the percent of women treated for DCIS (note the words "following treatment") who subsequently develop an invasive recurrence. Many studies have shown that women over-estimate this risk. Many studies have shown that recurrence rates are nowhere near 50% and that only 1/2 of all recurrences are invasive (the other half are DCIS). But remember - this is what happens after the DCIS is removed and treated; this has nothing to do with what might happen if the DCIS were not surgically removed (which is ultimately what this article is making recommendations on).   Second, when talking about how to reduce the anxiety of the women who over-estimate their risk, they present an option that might very well be successful but because this is not what they want to do, they use words that imply that this option will fail ("it's tempting to hope" and "it's entirely possible"). The fact is that nobody knows if better education of women and their doctors will be effective in reducing the anxiety levels of women who've been treated for DCIS. Maybe we should try it and see.

. 

"Because the "best guess" is that most DCIS won't progress to invasive cancer, the risk of overdiagnosis would be expected to be greater than 50%."
Translation: It's all just a guess. We really don't have a clue what % of DCIS will progress but we are using these numbers to make our point and support our case.

. 

The authors of this article and those who are involved with the studies discussed in this article have a laudable goal, which is to help women who don't need surgery avoid surgery, and "to identify nonsurgical means of treatment to prevent DCIS progression to invasive cancer." A wonderful goal. But if you carefully read what is being said, you realize that the support for their theory is built on quicksand. Or more accurately, it's built on conjecture. 

So back to a point I made in another post earlier today. I happen to agree that there is a lot of over-treatment of DCIS and I support the efforts that are being made to address this.  But I also worry that some doctors are jumping the gun. At some point in the future, when we have better information about the biology of DCIS, it may be possible to segregate DCIS into two different diagnoses, one that is a pre-cancer and is no longer classified as being "DCIS", and another that is an early stage breast cancer, and remains under the name "DCIS". I believe we will get to that. But we are not there now, and everyone knows that.  To suggest to patients today that some types of DCIS show be ignored and not diagnosed and/or not treated is, in my mind, irresponsible. It's playing with patients health and well-being, and maybe even with their lives.

dp4peace

Thanks for posting this articel maize!

"the trial essentially tested the value of finding the microscopic nonpalpable lesions of DCIS. The trial found no statistically significant difference in breast cancer mortality (and, in fact, mortality was nominally higher in the mammography group"

Donna (www.dcis411.com)