Fractionated stereotactic RT for resected single brain mets

[Kroll2013]

Dear colleagues,
I need your opinion concerning my pt
81 yo, in very good shape that presented with a single cerebellar brain met. The primary is colon ADK
This lesion was completly resected and the pt is planned for FSRT on the tumor bed .
What do you recommend as dose per fr and number of fr? PTV margin extension? Prescription IDL?
The treatment is LINAC-based .

Tx

---

Members don't see this ad.

 

[medgator]

NCCN still has adjuvant whole brain xrt after craniotomy as the category 1 rec.

I get the feeling few do that now though. Haven't had a case like this in awhile so I'm curious to see what people say

 

[Palex80]

I'd do 6 x 5 Gy. CTV = cavity + 1 cm, excluding natural borders. Add your PTV margin to that. Prescribe to the 95% isodose encompassing the PTV.

 

[Neuronix]

We do 5 Gy x 5 fx prescribed to volume. CTV/PTV expansion 3mm with Novalis/BrainLAB/ExacTrac. That involves the reinforced head mask with the infrared camera tracking system and floor mounted "cross-fire" x-rays for alignment. If you use a different IGRT setup you might wanna use larger margins.

Here's our single institution experience: http://link.springer.com/article/10.1007/s11060-014-1417-2

 

[Gfunk6]

I do the same as Neuronix, except I use cone beam CT.

 

[Kroll2013]

We do 5 Gy x 5 fx prescribed to volume. CTV/PTV expansion 3mm with Novalis/BrainLAB/ExacTrac. That involves the reinforced head mask with the infrared camera tracking system and floor mounted "cross-fire" x-rays for alignment. If you use a different IGRT setup you might wanna use larger margins.

Here's our single institution experience: http://link.springer.com/article/10.1007/s11060-014-1417-2

Thank you. I think i will stick to this.

 

[MegaVoltagePhoton]

Concur with Neuronix. I have a colleague that gets nervous and does 5x5 sometimes, but 6x5 is my standard.

As medgator pointed out, the level I evidence is for WBRT. However, I almost never use WBRT in these cases, unless the life expectancy is very poor and/or I know the patient won't follow up (i.e., for salvage SRS/SBRT).

 

[seper]

I do 15 Gy X 1 unless cavity is very poorly defined

 

[evilbooyaa]

Just out of curiosity, why do most of the folks int he thread not use single-shot SRS in a case like this? I get it if there is concern over toxicity to surrounding structures, but I'm imagining cerebellar lesions being relatively far away from most structures unless they happen to be right next to brainstem/spinal cord?

 

[ramsesthenice]

Just out of curiosity, why do most of the folks int he thread not use single-shot SRS in a case like this? I get it if there is concern over toxicity to surrounding structures, but I'm imagining cerebellar lesions being relatively far away from most structures unless they happen to be right next to brainstem/spinal cord?

Your right its about toxicity but usually not specific structures. Whole brain dose is the problem. Resection beds are large. Your V12 is almost always going to be well north of 10 cc in these cases. Based on old AVM data, and supported by more recent malignant literature, brain V10 and V12 are the more commonly utilized dosimetric parameters to predict risk of symptomatic radionecrosis.

 

[ramsesthenice]

Do note that V12 <10cc (or thereabouts) is only useful for single-fraction SRS, not fractionated SRS.

 

[Palex80]

What is your CTV? Cavity + ???

 

[ramsesthenice]

What is your CTV? Cavity + ???

Just cavity. Don't really do a CTV expansion unless there is a specific reason. I certainly don't cover pre-op edema for a met. Just a small PTV around the cavity.

Part of my experience is we have great but in for radio surgery by our neurosurgeons. They almost never take these out unless there is a dominant met or it's the only site of disease and they need tissue. Either way, you are usually talking about good sized tumors before surgery. Add post op changes and uncertainty on the MRI and the cavities are frequently pushing 3 cm or bigger. Too big for single fraction.

 

[medgator]

Just out of curiosity, why do most of the folks int he thread not use single-shot SRS in a case like this? I get it if there is concern over toxicity to surrounding structures, but I'm imagining cerebellar lesions being relatively far away from most structures unless they happen to be right next to brainstem/spinal cord?

What's the data for single fraction srs to a cavity? I'd say fractionated is standard practice, considering wbrt is still in the nccn guidelines as a level 1 recommendation.

Sometimes, you don't want to be too conformal in a post op vs intact met situation, and fractionated allows you to address bigger target volumes

 

[ramsesthenice]

I can tell you the only 2 patients I know it have hurt with SRS were post op patients with bigger cavities on a nationwide trial. They wanted specific doses based on cavity volume which were more than I liked and single fraction. One patient got 17 Gy with a V12 of 15 cc. She herniated two days later and died. The second got 17 Gy and had a v12 of 13 cc. She has since had a second operation for symptomatic necrosis without disease recurrence.

Smart guys wrote this trial and I try not to make too much of one persons limited experience. Still, the intact data was pretty clear that people started getting in trouble more frequently when they treated big volumes in a single fraction. We and others have seen very good control with fractionated SRS. I don't see the need to take chances with big volumes.

To Palexes point, if they are sending you people with small, well defined cavities, that is something different. But I never see that.

 

[seper]

can you guys please link data on V10 and V12 re: radionecrosis?

 

[evilbooyaa]

Thanks for the explanations. I didn't think about the larger size of CTV in a post-op sim vs an unresected met

 

[BobbyHeenan]

Good points all around in this thread.

As mentioned, target delineation for the post op cavity sometimes can be challenging. In training, we didn't technically add any additional margin for our single fraction radiosurgery post-op cases, but my attendings contoured a generous cavity. The Stanford single fraction experience suggested that a 2mm cavity to PTV margin showed improved local control. I suspect that generous contouring without margin is about equal to tight contouring plus 2mm margin.

I agree with ramsesthenice about those doses on the randomized trial - we have that trial open at my practice but I think the dose is too high so I don't enroll. Off study for gross total resections I favor 5 Gy X 5 to the cavity (frameless, cavity + 2mm = PTV) (if large), or 15 Gy X 1 (gamma knife, no margin, generous cavity contour) if the cavity is small.

V10 and V12 data is everywhere and very robust - from AVM treatment (Pitt/Flickenger/) and mets (Blonigan/Cincy - http://www.ncbi.nlm.nih.gov/pubmed/19783374). It does only apply to single fraction, so as mentioned a different metric is needed for fractionated SBRT approaches. I tend to look at things like gradient index as a surrogate for the moderate dose spill when evaluating plans, but volumetric dosimetry data correlating with radionecrosis is not well delineated in the fractionated brain SBRT literature.

 

[ramsesthenice]

The other relevant study that comes to mind is RTOG 90-05. Does not directly address V10 or 12 did clearly lay out that single fraction SRS gets trickier as the tumor volume increases. A lot of practices utilize goals that look pretty similar to those in that study.

 

[Palex80]

I am not sure that we should delineate only the cavity. As many of you have noted already, the cavity borders are not always clearly defined.
Furthermore I have concerns of tumor cell spread in the immediate area around the cavity (meningeal) due to the surgery. This is especially the case in lesions which are superficial.

The target volume definition is just not the same as with primary radiosurgery in my opinion.


I did a small series of patients with postOP RT only to the cavity + 1 cm. Out of 5 patients, two developed meningeosis carcinomatosa and died within a year. Probably bad luck, but it certainly made me worry.

 

[ramsesthenice]

I am not sure that we should delineate only the cavity. As many of you have noted already, the cavity borders are not always clearly defined.
Furthermore I have concerns of tumor cell spread in the immediate area around the cavity (meningeal) due to the surgery. This is especially the case in lesions which are superficial.

The target volume definition is just not the same as with primary radiosurgery in my opinion.


I did a small series of patients with postOP RT only to the cavity + 1 cm. Out of 5 patients, two developed meningeosis carcinomatosa and died within a year. Probably bad luck, but it certainly made me worry.

I think everyone agrees with you on this. Earlier when I said I don't do a CTV expansion I meant I don't just have some arbitrary isotropic expansion in mind. Just like you said these are hard. Post op changes etc. and no one really knows what the CTV should be when it looks like a bomb went off. Contours end up being quite generous compared to unresected patients and are often no where near symmetric. End up with pretty big targets which is why I almost always fractionate.

As for meningeal contamination I haven't seen anything like you have but it's not that far fetched to believe.

 

[Palex80]

I understand, thank your for the clarification. And I full agree with you that not fractionating and treating with doses common to radiosurgery a lot of brain tissue is not a good option.
Perhaps a randomized trial should be done for this question: WBRT vs. cavity RT for resected brain mets (1-2 mets).

 

[xrthopeful]

Perhaps a randomized trial should be done for this question: WBRT vs. cavity RT for resected brain mets (1-2 mets).

Already exists. The N107c trial

https://clinicaltrials.gov/ct2/show/NCT01372774

 

[ramsesthenice]

Already exists. The N107c trial

https://clinicaltrials.gov/ct2/show/NCT01372774

Yeah, there is a problem with that trial though. I love the principle of it but the fractionation schedule they use is weird. Not how most people practice and its hurt enrollment. Im worried the toxicity from SRS is going to look higher than it should...

 

[OTN]

I would do 35 Gy in 5 fx- a bit more dose, but recent data has suggested relative radioresistance of colorectal aca mets to SBRT/SRS. 2mm GTV to PTV margin.

I've been using this dose schedule for postop intracranial tumor bed SBRT with good results.

 

[radiaterMike]

I think everyone agrees with you on this. Earlier when I said I don't do a CTV expansion I meant I don't just have some arbitrary isotropic expansion in mind. Just like you said these are hard. Post op changes etc. and no one really knows what the CTV should be when it looks like a bomb went off. Contours end up being quite generous compared to unresected patients and are often no where near symmetric. End up with pretty big targets which is why I almost always fractionate.

As for meningeal contamination I haven't seen anything like you have but it's not that far fetched to believe.

There are data on leptomeningeal failure after srs to resection cavity with relatively high reported rates. It would be interesting to know these rates in the Patchell surgery alone arm to compare.

http://www.ncbi.nlm.nih.gov/m/pubmed/?term=radiosurgery+resection+leptomeningeal

 

[medgator]

There are data on leptomeningeal failure after srs to resection cavity with relatively high reported rates. It would be interesting to know these rates in the Patchell surgery alone arm to compare.

http://www.ncbi.nlm.nih.gov/m/pubmed/?term=radiosurgery+resection+leptomeningeal

This is why I am surprised how much whole brain xrt has been abandoned even in the post op setting.... This thread is likely a microcosm of overall community.

If someone presents with a big, edematous, midline-shift causing, symptomatic met that was resected, I'm not going to look up some protocol to treat their cavity....they're getting whole brain from me, still the level 1 recommendation from the NCCN

 

[thaddeus]

There are data on leptomeningeal failure after srs to resection cavity with relatively high reported rates. It would be interesting to know these rates in the Patchell surgery alone arm to compare.

http://www.ncbi.nlm.nih.gov/m/pubmed/?term=radiosurgery+resection+leptomeningeal

Which studies from this list report a high rate of LM failure? The two quotes from I found from those studies were 6% (Ling) and 14% (Ojerholm). Not sure these rates count as "relatively high", and for many patients would constitute an acceptable risk in exchange for avoiding WBRT. As with the WBRT vs SRS questions in general, I think you have to look at the whole patient picture (additional unresected mets, location, histology, systemic control, KPS, etc) rather than just making a blanket statement that all post-op patients should get WBRT per NCCN.

 

[medgator]

That's fine and dandy, but until we see level I data or at least a white paper from astro/acr on the matter, not just single institution data, the default for me in the community is still wbrt with cavity fsrt on a case-by-case basis

 

[radiaterMike]

Which studies from this list report a high rate of LM failure? The two quotes from I found from those studies were 6% (Ling) and 14% (Ojerholm). Not sure these rates count as "relatively high", and for many patients would constitute an acceptable risk in exchange for avoiding WBRT. As with the WBRT vs SRS questions in general, I think you have to look at the whole patient picture (additional unresected mets, location, histology, systemic control, KPS, etc) rather than just making a blanket statement that all post-op patients should get WBRT per NCCN.

Some others...

Strauss 10%
Ahmed 9%
Atalar 24% for breast and 9% for non breast

Though I didn't say these numbers mandate WBRT.

I agree it is an individual patient decision.

 

[xrthopeful]

The rates of LMD are exactly why we have the trial. It's going to help answer a very important question.

 

[radiaterMike]

recurrence patterns are indeed a secondary endpoint analysis of that trial.

Arguably survival and QOL/neurocognitive outcomes are of greater importance.

One would think that avoiding lepto would equate to improved survival and QOL. Although in a population of patients, does avoiding WBRT outweigh such potential benefit. Also- some patients will develop lepto after WBRT- perhaps more so in a surgical cohort.