Radiation after DCIS or not????

Radiation is used to try to destroy the tumor in some disease sites when the tumor is inoperable or difficult to remove.  To kill a big solid mass of cancer as opposed to smaller numbers of scattered cells they go to a much higher dose of radiation. 

redsox,

I've looked at mathteacher's profile.  She has no cancer info listed. Where did you see that she had a bilateral mastectomy.  I'm curious what she is basing all her comments on?

One thing to keep in mind is radiation technology has changed over the years. They can really pin point it now. Although there is still risk (heart, etc.) it is not as bad as it was years ago. Therefore I'm not sure how valid the stats are. We need to take into consideration that the long term risks might be reduced compared to 10 years ago.

Not sure if there are any studies on this.

barry, I just saw your question.  I agree with what redsox said about higher doses of radiation being required if you are trying to destroy a large mass (i.e the entire area of DCIS) vs. just a few random cells left after surgery, which is normally what radiation is used for.  The other thing to keep in mind is that radiation is about 50% effective.  If you have surgery and have a 12% recurrence risk after surgery, radiation will cut this risk to 6%, but it won't eliminate the risk completely.  So if you have the entire mass of DCIS, if it's high grade and/or has comedonecrosis and/or is multi-focal or multi-centric and/or if there is a large area of DCIS, then the risk that this DCIS will over time develop to become invasive cancer is very high, probably as high as 80% (or maybe even 100%).  So if radiation is only successful 50% of the time, this means that by not having surgery and having radiation treatment only, approx. 40% of the women will develop invasive cancer.  Obviously, that's not acceptable.

On the other hand, if someone has a very small single focus of low grade DCIS (let's say a 3mm tumor that's grade 1), the risk that this DCIS will eventually become invasive is probably quite low - maybe only 25%, maybe even less. In this type of situation, targetted radiation might be a good option instead of surgery.  I don't know if anyone is using this approach today but as medical science becomes better at identifying which types of DCIS are high risk to become invasive and which types are low risk, I could certainly see this approach come into use.

kickonit, thank you!  I'm glad my info has been helpful.

Mariekelly, after I responded to your post about the Fisher study, it occurred to me why they probably used the term "marginally significant" when referring to the difference in breast cancer mortality between the lumpectomy only group and the lumpectomy + radiation group.  Although the results in favor of lumpectomy + radiation are statistically significant at the 95% confidence level, the hazard ratio is only 0.82.  This means that lumpectomy + radiation provided an 18% benefit (in terms of reducing mortality) vs. lumpectomy only.  Within medical research, an 18% difference isn't very big.  When medical studies are done that look at risk factors, for example, they look for factors that double or triple risk; an 18% increase in risk, even if it's statistically significant, wouldn't be considered to be very important.  That's my guess on why the term "marginally significant" was used - they are referring to the importance of the results. That's judgemental, of course.  Personally I think an 18% reduction in mortality is pretty important.  Even the 6.5% difference in overall mortality between the lumpectomy + radiation and lumpectomy groups would be good result, if this directional difference were to be maintained with a larger sample (i.e. if the sample size was increased enough to make this result statistically significant at the 95% confidence level).  

This analysis of the Fisher study really points out how careful we have to be in interpreting - and sometimes reading too much into - research results.  The burden of proof is on the side that wants to show a difference; if results are being reported at a 95% confidence level, this means that they have to be 95% certain that any differences noted could not be randomly caused.  But if they are only 94% certain, the write-up of the study could say that "no differences were found".  This is why research studies tend not to focus on the items where "no differences were found" unless they have a large sample size (which would suggest that there really is no difference) or in the case of smaller samples, if the results are in fact very evenly split (50.3% vs. 49.7%, for example).

capqueen, I think that there is a benefit to avoiding a recurrence, even if that recurrence wouldn't have turned into mets. Avoiding chemo, avoiding Herceptin, those are good things.  So for me, radiation provides a benefit even if there is "no difference in overall survival".

But I also agree with you that when we are looking at mortality stats, the one that counts most is overall mortality.  Reducing the number of deaths from breast cancer isn't any good if the treatment itself (in the case of this discussion, radiation) leads to even more deaths.  But that's exactlly the point of everything I've written on this subject over the past few days.  The studies have not proven that there is no benefit from radiation in terms of reducing overall mortality.  The studies simply haven't proven, with 95% certainty, that there is a benefit to radiation.  The statements are similar but the implications are very different.  The Fisher study shows that radiation does provide a benefit in terms of reducing overall deaths; the problem is simply that this benefit has not been proven at a "statistically significant" level.  This doesn't mean that the result is meaningless or wrong; it may simply mean that the sample size wasn't large enough to drive a statistically significant result. As someone who has worked with research (non-medical) for 30 years, I've had to deal with small sample sizes a lot, and I've made lots of decisions based on "directional' data .  I'm not suggesting that we definitively conclude from the Fisher study that radiation does provide a benefit in terms of overall survival, but the results favor radiation and the authors of the study, in their own conclusions, continue to recommend radiation. 

The discussion in my previous post about whether or not the reduction in breast cancer deaths was significant was in response to a specific question about this from Mariekelly.  I was not suggesting that this metric was more important that overall survival.    

Kickon2it,  Curious, regarding what you said: I did mine in the prone position because I have asthma, and trusting what they were showing me - only a slight bit of rib was involved.  No heart, no lung.  Though I will always assume there was some impact there.  How were they able to show you?  I am amazed they were able to direct/target the radiation without othwith er organs getting in the way.  I am one of those people who got nuked 12 years ago.  I had DCIS majority was invasive.  Tumor was only 1.5, had lumpectomy, radiation and chemo(one-fits-all Adrimycin & Cytoxin).  So personally I think I was over-baked, so-to-speak.  I was 44 years old and as you all can relate "scared to death" thinking I needed to do something immediately.  What I know now, I wished I would have done much more research (or asked someone to help me with research) and I know I would not have chosen the radiation that was around 12 years ago.  But also, 12 years ago we were fighting tooth and nail trying to get our doctors to understand prophylactic choices and now since 1998 we have that choice and insurance will pay for it.  Back to the radiation I had - I can only hope that I will not be paying for having radiation (read: lung mets)  To all who are debating radiation or not - do your homework.  Know all your options before you make your decision.  Ask, ask, ask questions.

DebinATL - after my set up they were able to show me the trajectory of the beams and it appeared to bypass my heart and lungs but clip a couple of ribs.  Of course, I assume there has to be scatter and that even though the beams are targeted - some percentage must impact the area.  Because I was lying down and my breast was hanging beneath me, the trajectory of the beams appeared to have less impact on my chest cavity than when I was lying on my back.  Some argue that your heart flops forward in the position and it does you can see it in the simulation - but still the trajectory they showed me didn't cross it.  I might not be saying this well, but it was the computer simulation that is calculated to calibrate the angle of the beams.  They were able to show me a print out of it.

I will always wonder what the true impact will be on down the road....  But with my family history, I felt I had to go ahead with it.  Like we all know it is a personal decision made after much time spent agonizing over it!!!