NEJM Article : RT and Risk of Heart Disease

[runnerdoc7]

Interesting article headlining the NEJM this week. It is not available online yet, but in case anyone else subscribes in print I thought I would throw it out for discussion. Essentially the article establishes a clear relationship between breast RT and an increase in the rate of major cardiovascular events/death. Perhaps the most striking stats IMO were the % increase in rate of major coronary events per gray increase in mean heart dose. I'll just say that it is a lot. Finally, the increase in risk is proportional to mean heart dose.

Everyone should read the article themselves for a thorough understanding of the methods/results etc. At cursory first glance my thought was "oh no, another negative RT article in the NEJM," but after reading I was left wondering if this kind of data would not help justify expanding the use of more nuanced techniques for breast RT (FIF/IMRT/Breath Hold) to a broader subset of patients, especially those who already carry cardiovascular risk factors. If reducing mean heart dose by a gray can reduce risk as much as the article suggests, it may be a reasonable use of extra resources.

Either way, it is a well done study and a quick read; highly recommended. I would love to hear thoughts from others below.

---

Members don't see this ad.

 

[deleted4401]

I have never known the print version to come out ahead of the online version, especially for NEJM. Doing a Pubmed and Google Scholar search come up completely empty. Everyone knows there is a dose-response relationship, and that one of the main reasons why a benefit to breast cancer specific survival from RT hasn't been shown until recently is better technique. FiF is basically standard now for 3D cases (we do every case that way). IMRT doesn't bring the heart dose down, just improves homogeneity. People are using breath hold and prone positioning. The dose to the heart has never been lower.

"I'll just say that it is a lot." ... You think we are afraid of numbers?

Does anyone else know what article they are talking about?
S

 

[medgator]

Everyone knows there is a dose-response relationship, and that one of the main reasons why a benefit to breast cancer specific survival from RT hasn't been shown until recently is better technique. FiF is basically standard now for 3D cases (we do every case that way). IMRT doesn't bring the heart dose down, just improves homogeneity. People are using breath hold and prone positioning. The dose to the heart has never been lower.

Yup. And why looking at heart-dose literature from 20-30 years ago is so problematic. We know things like medial wedges etc. aren't good things to use yet much of the data we have is from that era (or even older).

And yes I have NO idea what article the OP is referring to. Very odd for it not be online but somehow in print.

 

[Gfunk6]

It's too early for April Fools Day.

 

[deleted4401]

Dude... Give an author name, title, and volume. Not vague descriptions of something that everyone past PGY-1 year knows and associated alarmist quotes. Articles do not come out in paper any more before being online. The Red Journal is nearly a year behind with what is "in press" and what is actually "pressed". NEJM is several months behind.

What does have to do with being cool? Everyone else "who is cool" gives a PMID, a link to the actual article, or a reference when wanting to have a discussion about a journal article. You've given none of the above.

EDIT: Thanks for posting the reference.

 

[RadRadRad]

Saw it too.

"Risk of Ischemic Heart Disease in Women after Radiotherapy for Breast Cancer"
Darby SC et al. March 14, 2013. NEJM

Uses patient data back to 1958...there is a methods section on how they calculated mean heart doses, but it would seem difficult to estimate from a plan from 1958 IMO.

They quote an increase risk of 7% per Gy of mean heart dose...with "no apparent threshhold"...guess they need to stop using fluoro to put in the cardiac stents.

 

[rymd]

A search for that author brought up this article http://www.ncbi.nlm.nih.gov/pubmed/23257897 titled "Radiation-related mortality from heart disease and lung cancer more than 20 years after radiotherapy for breast cancer." A SEER study from 1973-2008.

Solutions tend to lead to other problems. The challenge is that it'll take another 20 years to prove that our newer techniques are safer...

 

[Gfunk6]

Woah, take it easy everyone!

For future reference (pun intended), please do quote the source. In this particular instance, it does appear that the print version was released ahead of the electronic version, hence the confusion.

While I fully plan to read the article when it comes out, I'm pretty sure it will be another Rad Onc hatchet job by NEJM.

For what it's worth, I consider this paper by Larry Marks to be one of the most seminal papers on the subject. The title is "The incidence and functional consequences of RT-associated cardiac perfusion defects." They had 3D data and used the clear and meaningful endpoint of "cardiac perfusion deficits" as assessed by nuclear medicine studies. I use his V25 < 5% parameter as a strict dose constraint for all of our L sided breast plans. Fortunately, with modern technology the V25 is almost always < 1%.

 

[Neuronix]

At my institution, the breast surgeons are constantly pushing for less radiation, primarily due to cosmesis/reconstruction issues. I'm worried that publishing this data for a general audience per the NEJM will just lead to more negative popular press for radiation, leading to more refusals from referring docs and more patients declining radiation. For anyone reading from other specialties: these things have to be taken into context!

Certainly, keep dose as much off the heart as possible. That's a valuable message. But the British Columbia post-mastectomy RT study (including bilateral Co-60 straight into the bilateral IMNs and into the front of the heart!!!) at 20 year long term follow-up shows (http://www.ncbi.nlm.nih.gov/pubmed/15657341?dopt=Abstract):

XRT vs no XRT (n=318)
Overall survival 47% versus 37%
Cardiac deaths 1.8% versus 0.6% (not statistically significant)

Or the Danish 82b+c 15 year long term follow-up (1152 patients with 8+ axillary LNs removed, http://academicdepartments.musc.edu/surgery/education/journal_club/08-09/january09.4.pdf)

XRT vs no XRT (n=1152)
Overall survival 39% versus 29%
No difference in cardiac deaths at 10 years

Based on these two prospective phase 3 trials, I would not hesitate to provide radiation including to the IMNs with a known risk of dose to the heart. We use a constraint mean dose of 4Gy, and that is almost certainly equivalent to or less than the above mentioned trials. Why use a SEER database review or other retrospective data to justify changing your management when we have strong prospective data?

I know I'm preaching to the choir on this. I will make my own personal evaluation of the NEJM data when I can see it (why isn't it online???). But, this new data can easily be taken out of context in part by virtue of its being published in a high profile journal (Z11 anyone?), when we have large randomized long-term trials showing the benefit of chest wall and FULL regional nodal irradiation.

I suspect with the coming MA-20 (https://docs.google.com/viewer?a=v&...t5-9Ts&sig=AHIEtbRdTeAYzj7ctnJ0pYmbJbXqY-gW2A) we can have this debate again. i.e. Was it the SCV radiation or the SCV+IMN radiation that provided benefit? Does the IMN radiation outweigh the possible cardiac toxicity?

 

[Gfunk6]

I suspect that this will play out in a similar way to long-term androgen deprivation in men with high-risk prostate cancer. Despite decades of randomized data showing a both overall survival and cause specific survival advantages, there was a lot of hooplah about the risk of long-term androgen deprivation and cardiac toxicity. A recent meta-analysis put those fears to rest.

Similarly, meta-analyses of randomized breast cancer data shows an overall survival advantage to radiation at 15 years. This was done using older technology where the dose to the heart and ventricles was undoubtedly higher than it is today. Radiation saves lives, end of story.

Nevertheless, as Neuronix points out, we must remain politically savvy and engaged in this conversation lest our patients are put in jeopardy. Withholding XRT in women because of perceived cardiac toxicity is "penny wise, pound foolish."

 

[deleted4401]

Sorry about the harsh initial response.

But, I'm anything but anonymous and saying that's colleague would not be beneath me, as I can be quite ornery. Looking forward to reading their next hatchet job.

 

[seper]

Dr. Darby is the undisputed authority on late effects of unintentional cardiac irradiation. Can't wait to get my hands on the study.

 

[Palex80]

A search for that author brought up this article http://www.ncbi.nlm.nih.gov/pubmed/23257897 titled "Radiation-related mortality from heart disease and lung cancer more than 20 years after radiotherapy for breast cancer." A SEER study from 1973-2008.

Solutions tend to lead to other problems. The challenge is that it'll take another 20 years to prove that our newer techniques are safer...

So, the study basically says, that we shouldn't worry, if we irradiate our patients with post-1980-technology.
Well thank you for that important piece of information.
I will immediately stop irradiating my patients with 1970s technique.

I have to go now, sorry. I have to shut down my betatron ASAP.

 

[rymd]

It's here. www.nejm.org/doi/full/10.1056/NEJMoa1209825?query=featured_home

Interesting article headlining the NEJM this week. It is not available online yet, but in case anyone else subscribes in print I thought I would throw it out for discussion. Essentially the article establishes a clear relationship between breast RT and an increase in the rate of major cardiovascular events/death. Perhaps the most striking stats IMO were the % increase in rate of major coronary events per gray increase in mean heart dose. I'll just say that it is a lot. Finally, the increase in risk is proportional to mean heart dose.

Everyone should read the article themselves for a thorough understanding of the methods/results etc. At cursory first glance my thought was "oh no, another negative RT article in the NEJM," but after reading I was left wondering if this kind of data would not help justify expanding the use of more nuanced techniques for breast RT (FIF/IMRT/Breath Hold) to a broader subset of patients, especially those who already carry cardiovascular risk factors. If reducing mean heart dose by a gray can reduce risk as much as the article suggests, it may be a reasonable use of extra resources.

Either way, it is a well done study and a quick read; highly recommended. I would love to hear thoughts from others below.

 

[Gfunk6]

OK, read the paper a couple of times and here are my initial thoughts:

1. A patient population treated between 1958-2001 is not representative of the current state-of-the-art radiation technology.

2. The way they calculated mean dose to the heart and dose to the L anterior descending artery is highly suspect:

Individual radiotherapy charts, including a diagram or photograph of the treatment fields and a dose plan (where available) were copied. Virtual simulation and planning based on computed tomography (CT) (or, for a few regimens, manual planning) were used to reconstruct each radiotherapy regimen on the CT scan of a woman with typical anatomy.

I'm sorry but contouring a L anterior descending artery on a non-contrast CT is not an easy thing to do. I question using "a diagram or photograph" of treatment fields to do the same in an accurate manner. The necessarily had to presume "normal anatomy" which could not take into account things like cardiomegaly or other anatomic variations.

3. 78% of women in this study underwent mastectomy, so extrapolating the results to BCS is dubious.

4. No details on radiation. Tangents? Supraclav included? IMN included? PAB included?

4. If you choose to accept the conclusions of this study, then it is more of an indictment of MA-20 and the European trend of treating women with "comprehensive" radiation which will undoubtedly increase mean heart dose compared to straight tangents.

 

[radiaterMike]

one minor point Gfunk - the results of this study are not going to necessarily apply to your new consult with breast cancer, but the results are applicable to long term survivors of breast cancer, who may appear in your clinic (if you inherited patients).

The dosimetry in papers like this (M Stovall at MDACC has built a career using similar methods) are always suboptimal but certainly better than simply correlating outcomes to whether radiation was given. Any retrospective population-based study is going to have flaws related to lack of details. It is actually impressive what they did do with respect to dosimetry given the number of patients they analyzed.

I would simply interpret this as another (of many) studies correlating long-term cardiac risks with radiotherapy exposure, and (back to your new consult) use it to justify more advanced technology in the treatment of breast cancer. Anything that provides more ammo for rad. onc.s to justify IMRT to insurers is actually a good thing.

 

[deleted4401]

Sigh... Methodology and flaws of the article, the major issue is the convergence of factors that have led to the increase in the mastectomy rate in the United States as compared to other nations and the relative decline of BCS+RT. Some are good things, like BRCA testing and the (in my opinion, valid) need to protect against contra lateral malignancy. Also, good is that reconstruction is better - can use saline again, the use of TRAM flaps, and better onco plastic surgery in general. The bad factors are frustrating. The number one thing I see is the over utilization of MRI to make surgical decisions, which is expressly not recommended by the NCCN and most national organizations. Finally, the fear of radiation toxicity. There is an article in CNN today that discusses that most cancer centers have seen nearly a doubling of their mastectomy rates (they fail to mention MRIs).

The 25 year follow up to B06 has not shown an increase in cardiac mortality. We know with modern techniques that we can keep the dose to the heart to very low levels. I don't believe that the meta-analyses are wrong - I truly believe RT saves lives when done correctly. The problem with NEJM and the Times is that radiation oncology is in their cross hairs for economic reasons, and any chance to show that it may not work or that it causes unnecessary deaths will be prominently featured. How does this article get to the NEJM, but the EBCTG trial gets published in lower caliber journals? Do you think when the B06 cardiac mortality data gets published it will be in NEJM or a lesser, non-media viewed journal? I feel I know the answer already.

I hope there is an early and vigorous response to this article. The majority of the data says radiation helps, not hurts. This needs to be publicized. Otherwise, with MRIs and surgeons mentioning this data, we will take a backwards step in patient care.

S

 

[radoncapp2013]

The New York Times picked up this story:
http://www.nytimes.com/2013/03/14/health/radiation-modestly-raises-womens-heart-risks-study-says.html?hp

Sarah Darby, Silvia Formenti, and Louis Constine are all quoted.

 

[deleted4401]

Very interesting commentary. The quotes actually are reasonable and not alarmist. I think 5 Gy is really high, though, for mean dose. What are people using for their constraints? I'm aiming for V30 < 3 % and having no qualms blocking heart completely, if it's not blocking the lump cavity. Doing prone more, too. I still need to read article, but it actually doesn't sound poorly done, but it certainly isn't reflective of modern era. Also, has anyone mentioned that mastectomy patients (large part of study) tend to have worse disease and so they are more likely to get anthracyclines +/- RT?

Bet prone treatment picks up dramatically for left sided disease. Need a thread on that...

 

[Gfunk6]

In the NEJM article, a sizeable majority did not receive chemotherapy. For those that did, CMF was the standard of care at the time. In fact, more patients in the control group got chemo than the radiation cohort. One bias this study does not have is cardiotoxic modern chemo like doxorubicin or trastuzumab and the authors point this out.

 

[seper]

This is an excellent paper, as good as you can get with retrospective study incorporating DVH.
As discussed on this forum, there is resurgence of interest in PMRT for young women not meeting traditional ASCO PMRT indications. Dr. Darby's paper can cool that off a bit.

 

[medgator]

This is an excellent paper, as good as you can get with retrospective study incorporating DVH.
As discussed on this forum, there is resurgence of interest in PMRT for young women not meeting traditional ASCO PMRT indications. Dr. Darby's paper can cool that off a bit.

Perhaps just a bit. Still a very controversial area imo

 

[Neuronix]

This is an excellent paper, as good as you can get with retrospective study incorporating DVH.
As discussed on this forum, there is resurgence of interest in PMRT for young women not meeting traditional ASCO PMRT indications. Dr. Darby's paper can cool that off a bit.

I don't think this paper should cool off anything. From the discussion within the manuscript:

For a 50-year-old woman with no preexisting cardiac risk factors, radiotherapy involving a mean dose to the heart of 3 Gy would increase her risk of death from ischemic heart disease before the age of 80 years from 1.9% to 2.4% (i.e., an absolute increase of about 0.5 percentage points), and it would increase her risk of having at least one acute coronary event from 4.5% to 5.4% (i.e., an absolute increase of about 0.9 percentage points). If her mean cardiac dose were 10 Gy, her absolute risk of death from ischemic heart disease would increase from 1.9% to 3.4% (1.5 percentage points), and her absolute risk of having at least one acute coronary event would increase from 4.5% to 7.7% (3.2 percentage points).

Our constraint for full nodal radiation for left sided heart is 6 Gy. As per the above, that would imply about a 1% absolute risk increase in death related to heart disease and 1.5% increase in acute coronary event. You cannot execute a trial powered enough to show a 1% absolute risk difference, so I suspect this will never be seen in a single large trial. But, the survival benefits of radiation are clearly documented. I already cited British Columbia and Danish 82b+c which used full nodal radiation including IMNs. EBCTCG meta-analysis shows the 5% OS benefit for whole breast irradiation. Those survival numbers would include excess mortality for cardiac events. The excess mortality due to cardiac events is just insignificant compared to the benefits of radiation.

For PMRT not including IMNs, the mean heart dose varies, but let's say it's on the order of around 3 Gy. So what then? An absolute increase of cardiac mortality of 0.5%? It's practically insignificant. Yes, we shouldn't throw radiation around haphazardly, but we should continue looking for new indications that will benefit patients. The Chinese T1-2 N0 triple negative trial showed an OS benefit for PMRT of ~12% (http://www.ncbi.nlm.nih.gov/pubmed/21852010)!

Of course I'm basing my discussion on the PMRT group. For WBRT, the mean heart dose is often vanishingly small. I looked at a couple cases and sometimes the left sided mean gets as high as 2Gy (50Gy WBRT, supine, free breathing). That's a worst case scenario. For right sided that should be as low as tens of cGy. That presents almost insignificant risk to the patient IMO, though I have been counseling WBRT patients on a possible 1:1000 chance of excess cardiac mortality. Maybe I should increase that to 1:500. Prone positioning might in some patients get the lung and heart dose down a bit, but again by a clinically insignificant amount. Maybe it'd be worth it to buy the prone breast board for several 10s of thousands. I've been undecided on this for awhile.

In the end, I do not think this manuscript should change current standard of care management or trials including possible expansion of PMRT to other borderline risk groups.

The problem with NEJM and the Times is that radiation oncology is in their cross hairs for economic reasons

I don't think this is the real issue. I think the NEJM and the Times are both trying to increase readership to be competitive with other journals/newspapers, and in the NEJM's case trying to increase citations. The sensationalist (often negative) headlines get the most views and discussion.

 

[thesauce]

The major outlets that have picked up on this seem to have spun it favorably. That's good to see for a change.

 

[napoleondynamite]

Huh, yah Seper I think this trial is "excellent" in the same way HD6 was as a good historical paper to tell us what we already know - that STLI is toxic and not the standard of care. Unfortunately, medoncs and the general public don't typically grasp the nuances of STLI vs IFRT or INRT..they just see NEJM publishing something in 2012 that shows RT in lymphoma = higher risk of death (from drowning..). Same with this study..it may be a nice historical paper telling us that old RT techniques have a high risk of cardiac toxicity..but how is that relevant to current practice?? The answer is, it really isn't. It's more a political agenda that NEJM seems bent on pushing in recent years that is anti-RT. And just like when HD6 was published we will probably have to defend this at every breast tumor board for awhile, which is a waste of time and resources from what we should actually be discussing.

 

[deleted4401]

Upvote

Huh, yah Seper I think this trial is "excellent" in the same way HD6 was as a good historical paper to tell us what we already know - that STLI is toxic and not the standard of care. Unfortunately, medoncs and the general public don't typically grasp the nuances of STLI vs IFRT or INRT..they just see NEJM publishing something in 2012 that shows RT in lymphoma = higher risk of death (from drowning..). Same with this study..it may be a nice historical paper telling us that old RT techniques have a high risk of cardiac toxicity..but how is that relevant to current practice?? The answer is, it really isn't. It's more a political agenda that NEJM seems bent on pushing in recent years that is anti-RT. And just like when HD6 was published we will probably have to defend this at every breast tumor board for awhile, which is a waste of time and resources from what we should actually be discussing.

 

[Gfunk6]

Official talking points from ASTRO:

1. The benefits of radiation therapy outweigh the risks of cardiac toxicity in the majority of patients for whom radiation therapy is indicated. Current technologies continue to advance and provide many ways to protect the heart, particularly when treating the left breast.
2. The NEJM study is a significant retrospective look at treatment from 1958-2001, but it is an analysis of patient records and is estimating the risks of radiation exposure to the heart during this time interval when technology was not as advanced as it is today.
3. While the study indicates that there is increased cardiac risk for those patients who received radiation, it is important to keep in mind the significant advances in technology from 1958 to 2001. In addition, much has changed just since 2001. It is also important to note that even with older technologies and slightly increased risk of cardiac events, there is a substantial and significant overall survival and breast cancer mortality benefit with radiation in appropriately treated breast cancer patients.
4. As with any medical intervention there remain risks and benefits to treatment. With modern technology, the risk of cardiac exposure can be minimized to levels far below those achieved in previous decades. It has clearly been shown that in patients treated by lumpectomy as well as in patients treated by mastectomy for whom radiation is indicated, the benefits of radiation, in terms of overall survival and breast cancer mortality, far exceed the potential risks.

 

[RadOncDoc21]

Huh, yah Seper I think this trial is "excellent" in the same way HD6 was as a good historical paper to tell us what we already know - that STLI is toxic and not the standard of care. Unfortunately, medoncs and the general public don't typically grasp the nuances of STLI vs IFRT or INRT..they just see NEJM publishing something in 2012 that shows RT in lymphoma = higher risk of death (from drowning..). Same with this study..it may be a nice historical paper telling us that old RT techniques have a high risk of cardiac toxicity..but how is that relevant to current practice?? The answer is, it really isn't. It's more a political agenda that NEJM seems bent on pushing in recent years that is anti-RT. And just like when HD6 was published we will probably have to defend this at every breast tumor board for awhile, which is a waste of time and resources from what we should actually be discussing.

Amen!

 

[Seldon1985]

The major outlets that have picked up on this seem to have spun it favorably. That's good to see for a change.

I've seen that too... I talked to my chief of breast rad onc today and that concept of "radiation to the heart is bad" is a good one (albeit one that we knew). Although the methodological flaws abound in this paper, what we currently do in the clinic will still go on - do your best to minimize dose to the heart.

 

[seper]

What the hell does anything have to do with Vietnam?

Huh, yah Seper I think this trial is "excellent" in the same way HD6 was as a good historical paper to tell us what we already know - that STLI is toxic and not the standard of care. Unfortunately, medoncs and the general public don't typically grasp the nuances of STLI vs IFRT or INRT..they just see NEJM publishing something in 2012 that shows RT in lymphoma = higher risk of death (from drowning..). Same with this study..it may be a nice historical paper telling us that old RT techniques have a high risk of cardiac toxicity..but how is that relevant to current practice?? The answer is, it really isn't. It's more a political agenda that NEJM seems bent on pushing in recent years that is anti-RT. And just like when HD6 was published we will probably have to defend this at every breast tumor board for awhile, which is a waste of time and resources from what we should actually be discussing.

 

[seper]

Most people in this thread, presumably, recent graduates, picked up on "obsolete" techniques described in this paper. I trust that you block heart apex completely.

It not the case in real practice. My senior partners continue to happily blast away at chest wall and IM nodes using straight-on mixed electron-photon beams or high energy electrons alone. I know a busy RadOnc who still treats IM nodes for all node+ women. Mean heart dose routinely reaches 7 Gy in plans I've looked at.

 

[medgator]

Most people in this thread, presumably, recent graduates, picked up on "obsolete" techniques described in this paper. I trust that you block heart apex completely,

It not the case in real practice. My senior partners continue to happily blast away at chest wall and IM nodes uses straight-on mixed electron-photon beams or high energy electrons alone. I know a busy RadOnc who still treats IM nodes for all node+ women. Mean heart dose routinely reaches 7 Gy in plans I've looked at.

you're a product of your training. I think your examples are the kind of thing that lead the adoption of time-limited certificates in rad onc (as well as many other specialties). I still know plenty of old-times who treat definitive H&N and lung at 180 a day with chemo (despite NCCN expressedly mentioning 2 Gy/day as the SOC, at least in lung)

 

[Pewl]

What the hell does anything have to do with Vietnam?

The dude abides...