In today's Wall Street Journal: Breast IORT vs. EBRT

[Gfunk6]

http://www.wsj.com/articles/alterna...stage-breast-cancer-with-radiation-1440448587

Another radiation article in the mainstream media. If you are not already aware, there was a MAJOR controversy in the Red Journal over the statistical methods used in the TARGiT publication. There were quite a few 'letters to the editor' explaining this.

---

Members don't see this ad.

 

[BobbyHeenan]

I'm in agreement that more follow up is needed, but enterprising hospitals are going to jump on this data and you/we as rad oncs need to be familiar with the trial and the concepts. My surg oncs and breast surgeons are asking me about it already. It will be interesting to see the follow up and how patterns of recurrence evolve in addition to seeing longer term follow up on cosmesis for those patients that had IORT combined with XRT.

In addition, proton therapy is listed in the side bar of the article as an option. Why this is being done for right sided (or even left sided) whole breast cases is unclear to me - I see it done all the time off study or on a "registry study," but it happens and I think it's really skirting the whole cost issue to discuss cost in the setting of breast radiation and not even mention cost for proton therapy.

I personally have been a very big fan of hypofractionation for whole breast radiation. I am aggressive with heart blocking and am getting more discerning with boost, saving it for high grade, very close margins, young patients, triple negative, etc. So when I talk to my patients about APBI (we do brachy for this) versus canadian fractionation, they seem to lean toward 3 weeks rather than BID 1 week. IORT probably has a role (may be especially good for those older ER positive patients we sometimes observe anyway), but I'm not ready to tell my hospital administrators we need a machine for it STAT. Maybe we'll get left behind, but the data needs a few more years I think.

 

[CharlestonRadOnc]

@BobbyHeenan you are not going to be left behind... I completely agree with you.

For low-risk invasive & DCIS patients I think that hypofractionation is the way to go. IORT and brachytherapy are overkill. We follow the Harvard fractionation scheme of 265 cGy x 16 followed by a boost 250 cGy x 4 for a total of 4 weeks.

 

[seper]

no data to support 3 weeks tx in DCIS, right?

 

[BobbyHeenan]

no data to support 3 weeks tx in DCIS, right?

Not direct phase III data, correct.

But there's no mastectomy vs breast conserving data for DCIS either, and we think it's OK to extrapolate there so I think it's reasonable to extrapolate for DCIS. It's my opinion here, but there are so many more important trials that should be done that the costs and time of doing a phase III trial for DCIS for this question (hypo vs standard fractionation) aren't worth it in my mind.

There's such a push to eliminate XRT from DCIS going on (see previous NYT article, increasing mastectomy/reconstruction rates), we're doing ourselves no favors by insisting on 6 weeks of treatment. I think it's perfectly fine to offer standard fractionation, but for patients that can understand nuance and extrapolation, I personally also offer 16-20 fractions like CharlestonRadOnc mentions above for DCIS.

 

[medgator]

I use the retrospective data from Canada to offer it to DCIS pts. I am hesitant to do it in high-grade pts without adding a boost or just switching to conventional XRT with a boost. I also get nervous with patients on the younger side (I also am hesitant to hypofractionate anyone under 60 years of age, even though ASTRO advocates age 50 as a cutoff). I also tell them the data is not as strong because there is no phase III RCT data, but it's probably "OK" based on the retrospective stuff.

http://www.redjournal.org/article/S0360-3016(14)03533-0/abstract

 

[BobbyHeenan]

I use the retrospective data from Canada to offer it to DCIS pts. I am hesitant to do it in high-grade pts without adding a boost or just switching to conventional XRT with a boost. I also get nervous with patients on the younger side (I also am hesitant to hypofractionate anyone under 60 years of age, even though ASTRO advocates age 50 as a cutoff). I also tell them the data is not as strong because there is no phase III RCT data, but it's probably "OK" based on the retrospective stuff.

http://www.redjournal.org/article/S0360-3016(14)03533-0/abstract

I think this is very reasonable.

I get a little more "uncomfortable" with the hypofrac when I feel like a boost is needed (like you mention above, high grade DCIS, younger patient, etc). I still use boost sometimes in hypofrac (like mentioned above 2.5 Gy X 4 - I believer Harvard, UPMC, UAB, and Haffty/RWJ do this as well per ASTRO talks and small papers), but now we're getting even further away from really strong data, so I try to be judicious about boost and push for more standard fractionation in younger DCIS patients that I think need a boost.

There's a lot at play here and lots of reasonable opinions. I do tend to think that just saying across the board all DCIS needs standard fractionation because no Phase III data is probably too dogmatic for me, though some colleagues I respect a lot are still sticking to that paradigm.

 

[CharlestonRadOnc]

The colleagues that you respect who are sticking to a conservative paradigm on DCIS are shooting our field in the proverbial foot. We have phase III randomized data that HFRT is safe and effective for low-grade invasive cancer but we are "uncomfortable" using it for a non-invasive process? Many thought leaders including Gary Freedman from UPenn are treating DCIS with HFRT... it's a reasonable thing to do. Yes - whether or not to boost is very controversial and there is no right answer.

There are many treatment practices that are going on right now without Phase III data....
SBRT to a peripheral primary and chemoRT to mediastinal nodes for Stage IIIA/B NSCLC
Hormones + RT for prostate cancer patients s/p RT with biochemical failure

Unfortunately we do not have Phase III randomized data for everything... and while we're waiting for this data we're doing miserably against a wide variety of tumors.... it's a balancing act between moving the field forward and improving patient outcomes while minimizing potential harm to our patients.

 

[medgator]

"SSBRT to a peripheral primary and chemoRT to mediastinal nodes for Stage IIIA/B NSCLC"

anyone really doing this outside of a trial or academic setting?

 

[CharlestonRadOnc]

@medgator the answer to your question is yes. Also - many institutions are treating borderline/unresectable pancreatic CA patients with SBRT off-trial. Personally I'm not a big fan of SBRT for pancreas... but you get the general idea.

 

[medgator]

@medgator the answer to your question is yes. Also - many institutions are treating borderline/unresectable pancreatic CA patients with SBRT off-trial. Personally I'm not a big fan of SBRT for pancreas... but you get the general idea.

I think doing hypofx off protocol in the community for dcis is probably not as big of a leap as doing sbrt for stage III nsclc or pancreatic ca. Where is the phase III or large single-institution experience for that?

 

[Palex80]

IORT and brachytherapy are great, but in my opinion we are missing probably the easiest way to fo partial breast irradiation = external beam partial breast irradiation.
The RTOG 10x3.85 Gy bid schedule was clearly an overkill, delivering way too much dose in short time.

I use the START B fractionation schedule for low-risk patients:
15 x 2.66 Gy on the cavity/postoperative changes with a 2 cm margin.
Works like a charm so far.

 

[BobbyHeenan]

IORT and brachytherapy are great, but in my opinion we are missing probably the easiest way to fo partial breast irradiation = external beam partial breast irradiation.
The RTOG 10x3.85 Gy bid schedule was clearly an overkill, delivering way too much dose in short time.

I use the START B fractionation schedule for low-risk patients:
15 x 2.66 Gy on the cavity/postoperative changes with a 2 cm margin.
Works like a charm so far.

I agree that if we had a good way to do external beam APBI quickly it would be the best way to go about it. But, as you said, the cosmesis on the RTOG study for external beam in my experience (and what I've seen reported) has been sub-optimal for external beam. Some prospective data with protons and photons suggested that a slightly attenuated external beam dose as compared with the RTOG trial (32 Gy in 8 fractions, BID) had good cosmesis for photons, but poorer for protons - http://www.redjournal.org/article/S0360-3016(14)00446-5/abstract)...so maybe we've still got a ways to go to figure out appropriate dose/fractionation.

I haven't seen the cosmesis issues with APBI brachy, so when our group does APBI now we use brachy only.

 

[arthurdent]

Not long term follow-up, but DCIS patients included on this randomized MDACC study:
http://www.ncbi.nlm.nih.gov/pubmed/26247543

Also, DCIS is eligible for enrollment on RTOG 10-05.

The ASTRO guidelines on hypofrac need to be updated....

 

[RadRadRad]

I agree that if we had a good way to do external beam APBI quickly it would be the best way to go about it. But, as you said, the cosmesis on the RTOG study for external beam in my experience (and what I've seen reported) has been sub-optimal for external beam. Some prospective data with protons and photons suggested that a slightly attenuated external beam dose as compared with the RTOG trial (32 Gy in 8 fractions, BID) had good cosmesis for photons, but poorer for protons - http://www.redjournal.org/article/S0360-3016(14)00446-5/abstract)...so maybe we've still got a ways to go to figure out appropriate dose/fractionation.

I haven't seen the cosmesis issues with APBI brachy, so when our group does APBI now we use brachy only.

Livi et al. Randomized. 500 pts. Whole breast 50 in 25 vs 30gy in 5 fx external beam partial breast given q.o.d. 5 yrs follow up. Equal control and less side effects with 30/5
btw, allowed dcis although only about 15% of study population

 

[hot sauce]

RTOG 9804 was amended to allowed hypofx. However, I do not recall if they reported what percentage of patients received hypofx. As was mentioned earlier, we don't have strong phase III evidence for breast conservation in DCIS if you do not wish to extrapolate from invasive data. I guess we see the entire spectrum with some rad oncs happy to jump on techniques like IORT with questionable data while others drag their feet for hypofx despite the strong data.

I am not trying to call out anyone in particular. At my own institution there are some physicians who continue to treat the vast majority of their patients (DCIS and invasive) with standard fractionated treatment. If this pattern of practice continues nationally I suspect that we soon will have payers refusing to pay for standard fx; I would rather the choice remain with the physician but if we don't use resources prudently we will likely lose the choice.

 

[Neuronix]

I think doing hypofx off protocol in the community for dcis is probably not as big of a leap as doing sbrt for stage III nsclc or pancreatic ca. Where is the phase III or large single-institution experience for that?

For pancreas SBRT, our institution recently published a 159 patient series. We've been doing it since 2008 for borderline resectable and some locally advanced. Very low toxicity observed, no increased surgical complications for the ones who go to surgery. The technique is relatively complicated and requires some care, however.

 

[medgator]

For pancreas SBRT, our institution recently published a 159 patient series. We've been doing it since 2008 for borderline resectable and some locally advanced. Very low toxicity observed, no increased surgical complications for the ones who go to surgery. The technique is relatively complicated and requires some care, however.

Is that the single-fraction 20+ Gy protocol? Your last sentence is key.... I think I'd be very careful doing it outside of an academic setting.

 

[Neuronix]

Thanks for asking. No I think most groups are going to 5 fraction pancreas SBRT these days.

Here's the paper I'm referring to:

http://www.ncbi.nlm.nih.gov/pubmed/25734581

It's an update of this one:

http://www.ncbi.nlm.nih.gov/pubmed/23562768

 

[napoleondynamite]

I transitioned to SBRT for pancreas several years ago based on that Moffit data. As long as you keep the 35-40Gy volume tight, it's a safe technique IMO. I contour the gross disease in the pancreatic head, hit as much of that as I can with the 35-40Gy then the remainder of the pancreatic head to 25Gy, which allows meeting the duodenal and small bowel constraints fairly easily. I've probably treated 25-30 patients and I have not yet seen any complications.

I think SBRT for locally advanced makes a lot of sense. I've treated several borderline resectables as well with good results.

 

[RadOncDoc21]

Results good as in local control, survival, and toxicity?... compared to everything else (which all give a MS of about 12 months) with some pretty toxic regimens. Just asking because I have no idea.

 

[thesauce]

I transitioned to SBRT for pancreas several years ago based on that Moffit data. As long as you keep the 35-40Gy volume tight, it's a safe technique IMO. I contour the gross disease in the pancreatic head, hit as much of that as I can with the 35-40Gy then the remainder of the pancreatic head to 25Gy, which allows meeting the duodenal and small bowel constraints fairly easily. I've probably treated 25-30 patients and I have not yet seen any complications.

I think SBRT for locally advanced makes a lot of sense. I've treated several borderline resectables as well with good results.

Makes sense. What are you expansions on the GTV?

 

[napoleondynamite]

Results good as in local control, survival, and toxicity?... compared to everything else (which all give a MS of about 12 months) with some pretty toxic regimens. Just asking because I have no idea.

Results good in terms of R0 resection after SBRT - Moffit reported I think 90% R0 resection after SBRT

 

[napoleondynamite]

Makes sense. What are you expansions on the GTV?

I can't say I really expand the GTV. I contour the largest area that I feel I can safely get the dose to..then I give what I would consider the surrounding CTV the lower dose of 25Gy, which usually includes the entire pancreatic head.

Some of this is admittedly made up. But I have seen at least as good of palliation as with 30/10 and certainly less toxic than putting a patient through 5.5 weeks of RT, or worse, chemoRT based on the crappy ECOG data or whatever. I just don't believe in that for these patients with a terminal disease...even if that data is correct and treatment buys them 2 months OS benefit, those 2 months are usually not high quality months. I discuss all of that with my patients, most choose SBRT.

 

[seper]

30 patients? You should write it up as an abstract.

 

[napoleondynamite]

30 patients? You should write it up as an abstract.

I would, but at least 2/3 were treated at my prior job, so I no longer have access to the data.

Just sharing my own personal experience, take or leave it.

 

[Neuronix]

Makes sense. What are you expansions on the GTV?

We focus GTV on the gross disease based on EUS report and EUS implanted fiducials, PET/CT, panc protocol CT (+/- MRI), and sim CT (3D without then 4D with small amount of oral contrast). We have a fluoro unit that we use to decide about whether to use ITV or gating (1cm sup-inf motion cutoff).

That GTV then goes to 30 Gy with 3-5mm PTV expansion. Then there's a high dose GTV selected, which for borderlines is the tumor abutting the vessels. Dose limits are on the organs at risk as specified in the papers. Organs that are close to the PTV (typically duodenum and/or stomach) are typically contoured with motion to make a PRV. Sometimes you have to reduce the dose minimum a bit around the OARs to get the OAR doses in tolerance.

Treatments are done with daily CBCT and setup with the Varian RPM system.

My personal belief is that if we hit the tumors as hard as possible, we will get more patients to resection and possibly increase cure rates. For locally advanced tumors that are at least stable on FOLFIRINOX, we're getting more and more through margin negative resection with these aggressive techniques. For borderlines, we get about 50% through the whole protocol with margin negative resection. That's pretty good considering 10 years ago most institutions would have never operated on them at all! Certainly we get this done quickly, with minimal toxicity, and local control is high even for the ones we can't save.

Anyway, I need a job, so PM me and I'll come do this with you