What would you do?

[napoleondynamite]

---

Members do not see this ad.

SCLC previously treated to mediastinum as well as PCI. Now presents with mets and cord compression with responsible met within the prior treated field. Previous cord dose is already just under 45Gy. Pt has urinary retention, otherwise intact. The met appears intradural, not extending from the vertebral body.

Options:

-Go ahead with XRT. Paralysis risk from XRT vs paralysis risk from tumor, pray the odds are with you.
-Possible role for intrathecal chemo?
-Likely not a neurosurg player as the met is intradural..
-Steroids with no further tx?

Just curious what others would do in this situation.

 

[deleted4401]

Maybe important to know how long ago XRT to mediastinum, but with recurrent SCLC, may not be an issue.

Start with steroids, see if there is any improvement.

With recurrence, probably dead soon (MS - few months), late toxicity not too high on list of worries.

I'd treat 3-5 fractions of 2.5 Gy, if most recent RT >9-12 months ago. SCLC should melt right away.

If <9-12 months, would tell them very high risk of toxicity, but if patient/family cool with that, would treat. If not okay, BSC with 'roids sounds solid.

IT chemo sounds horrible.

Surgery sounds horrible.

If patient was just having pain, and no acute motor/urinary/bowel sx, I'd consider SBRT. Have treated a few with good results ...

-S

 

[clinonc]

In the UK we'd be likely to give a single 6-8 Gy fraction here.

http://www.rcr.ac.uk/docs/oncology/pdf/DoseFract_417Spinal.pdf

See ref 8.

 

[IndyXRT]

IT chemo tends not to do well at shrinking bulk disease. Systemic chemo would be a better option, and might be considered before re-irradiating (small round blue cell tumors with cord compression can be treated with systemic chemo).

 

[napoleondynamite]

Only 4 months has passed since the patients prior XRT to the mediastinum. Still, even with the short timeframe, the risk of toxicity to the spinal cord should be relatively low, say 10-20%? SimulD would you still be inclined to offer a few fractions after a frank discussion with the family, or does the timeframe scare you away? Just curious. Thanks for the interesting discussion.

 

[OTN]

You could do SBRT- U. Pitt has a series of pts treated c SBRT who previously received EBRT and presented c recurrence in-field. Our institution (WashU) currently has a stereotactic dose-escalation protocol for pts presenting c disease recurrence in-field after receiving EBRT.

There's not a lot of data on dose (their series ranged from 12 to 25 Gy x 1), but something like 10-14 Gy x 1 would probably be fine.

 

[Gfunk6]

You could do SBRT- U. Pitt has a series of pts treated c SBRT who previously received EBRT and presented c recurrence in-field. Our institution (WashU) currently has a stereotactic dose-escalation protocol for pts presenting c disease recurrence in-field after receiving EBRT.

I agree, @ my institution we would do CyberKnife (SBRT). Whenever we have previous XRT to the spine (near/at dose tolerance) SBRT can deliver a therapeutically meaningful dose while reducing the risk of complications.

You can possibly use IMRT but it would be more difficult.

With this patient the only goal is to hold off cord compression for as long as possible. Agree with SimulD that chemo will likely do nothing and surgery would be morbid (though I would want a Neurosurgeon to sign off).

 

[fettucine]

I agree, @ my institution we would do CyberKnife (SBRT). Whenever we have previous XRT to the spine (near/at dose tolerance) SBRT can deliver a therapeutically meaningful dose while reducing the risk of complications.

You can possibly use IMRT but it would be more difficult.

With this patient the only goal is to hold off cord compression for as long as possible. Agree with SimulD that chemo will likely do nothing and surgery would be morbid (though I would want a Neurosurgeon to sign off).

I Second SBRT 24-30Gy, recent article in red-journal (one of many recently) with safety report in low number of patients. I don't think doing nothing or steroid is an option, the patient will likely progress to paralysis very quickly, you may offer a chance to delay this with RT. RTOG (at some point) will be coming up with spinal SBRT trial, it has been in works for years.

Perhaps out of the box thinking is appropriate for this hence my argument below:
You have to wonder that these maybe clonogenically resistant cells, and therefore therapeutic gain would be low with "conventional" RT. With this argument even 1-2 fractions of low dose neutrons (U of Washington) maybe considered in combination with photons. We have actually tested these clonogenic resistant cells with High LET radiation (albeit in-vitro) with promising results. Consider Protons as well (not LET, but merely conformality) depending on how much the patient's family is willing to travel.

So in order of preferability, reccom: 1. SBRT or Protons 2. Neutrons + take your pick photon modality. Let us know what you decided.

cheers,
-f8

 

[clintpark]

I would advise against SBRT for two, maybe three, reasons.

If the patient already has cord compression, by definition, it's an urgent treatment. The attention to details and QA required for SBRT is not compatible with it.

Again, with pre-existing cord compression, it's already pressing hard on the cord, it's going to be very difficult to treat (even if we don't give any margin) without risking more injury to spinal cord. If SBRT does what we want and the tumor shrinks, then we are blasting the cord with full dose (assuming you do more than one fx treatment)... I'm not comfortable. Simul mentioned that the long term toxicity probably doesn't matter too much, I don't want to exacerbate the situation by giving him an acute toxicity.

Spine SBRT is not the first case of SBRT you want to do. If your institution has experience with treating other easier sites with SBRT, then I would consider it for non urgent, vertebral body mets without cord compression.

 

[kepler2]

fettucine-

you're dealing with cord compression in the setting of previously irradiated cord. If you're suggesting using neutrons to treat in this situation, what RBE are you assuming for neutron effects on neural tissue? Are there data to support that the RBEtumor:RBEcord ratio is >1 in this situation?

Also, in the setting of cord compression, the conformality that can be achieved with tight CTV-PTV margins (stereotactic setup) with a good IMRT plan will easily rival that of patched protons. The main benefit with protons would be in reduced integral dose to (non-cord) normal tissues.

 

[deleted4401]

I think I said that if it was non-urgent, SBRT was an option for an intradural met. We do it frequently at UPMC (someone mentioned the paper below). Another very famous doctor there (you've probably seen him on telesion, he's devilishly handsome and has quite a personality) submitted a case report of SBRT for intradural met and a review of the literature, and it should be published soon.

But urgently, I think my original idea of a few fractions for SCLC would probably be more logical, assuming we keep total BED under 120ish and each individual course <95ish. It should respond. Should be relatively low toxicity, 10-20% sounds fair to quote (I'd imagine it's lower, but highballing is always a good idea, when it comes to tox). 4 months is short time, you want 6 months usually, but with median survival ranging 2-3 month range, may not matter.

What did you do?
S

 

[Palex80]

I had a similar patient with a NSCLC recurrence. He was treated in 2006 with 50 Gy mediastinal after undergoing operation and chemotherapy for a Stage IIIb NSCLC.
He developed a recurrent tumour in the paravertebral space this spring, which destroyed the Th6 vertrebral body and invaded intraspinally. Multiple small lung mets were also diagnosed.
He showed no signs of neurologic deficit, so the surgeons turned down any kind of operation. He did suffer from pain. The medical oncologists started Pemetrexed + Carboplatin and asked us what we wanted to do.

After reviewing his plans, we saw that the 80% isodose went straight through his spinal cord at thei height of Th6, this means he had about 40 gy at 1,6 Gy/d delivered there in 2006.

Taking into account the Nieder calculations

Int. J. Radiation Oncology Biol. Phys., Vol. 61, No. 3, pp. 851&#8211;855, 2005

and
Int. J. Radiation Oncology Biol. Phys., Vol. 66, No. 5, pp. 1446&#8211;1449, 2006​

​

​

​

we calculated that we could probably retreat him safely. He finished treatment last week without any difficulties. Pain was already going back.
We gave him 36 Gy at 2 Gy/d.​

This totals to 144 Gy BED, which translates into 3 points. Thus he is at "Low Risk" with a 3% risk for myelopathy.
This is the maximum we would try. However one major benefit in our patient was that he finished treatment 3 years ago, allowing for some cord healing to take place in the mean time.​

 

[napoleondynamite]

Thanks everyone for the input, this has been a good discussion!

We don't do a lot of SBRT at my institution (unfortunately). My attending had no interest in re-treating, the pt has been sitting on the floor on steroids, luckily no change in neuro sx.

Surprisingly, neurosurg thinks he's a player and he is headed to the OR tomorrow 😱

Crossing my fingers..

 

[OTN]

Thanks everyone for the input, this has been a good discussion!

We don't do a lot of SBRT at my institution (unfortunately). My attending had no interest in re-treating, the pt has been sitting on the floor on steroids, luckily no change in neuro sx.

Surprisingly, neurosurg thinks he's a player and he is headed to the OR tomorrow 😱

Crossing my fingers..

How about THAT? Good for the patient- by far the best option.

Interestingly, Pitt treats (and I believe has published on) spinal cord compression with SBRT. Without the experience/staff (dosimetry, really) necessary to do this quickly, however, I agree that it would not be the best case to jump into the stereotactic pool with.

Good luck to the pt- we all hope he does well.

t

 

[Gfunk6]

Agree that Neurosurgery was the best option as long as they were willing to operate. Our institution published a somewhat relevant retrospective study in the Red Journal in the July 1 issue looking at SBRT for spinal mets in patients with and without prior irradiation (PMID: 19095374). Basically there was no difference in OS or PFS and no cases in either group of spinal cord myelopathy or radiculopathy. Usual caveats of a retrospective study apply of course, but it is a decent prelim study.

 

[napoleondynamite]

Really? Having worked with Dr. Gerszten for the past 2 months I found this hard to believe. So I asked him. It hasn't been done at Pitt, but has been done at henry ford and presented in abstract form. The case you may be referring to was an intradural lesion, which is different from cord compression.

The case I mentioned WAS intradural (see first post on the thread). I think that was probably what OTN was referring to.

 

[MOHS_01]

Not to hijack the thread, but I am envious of the good discussion here in this thread. These discussions, on occasion, make me regret not looking into radonc futher. I was really interested in it after a summer program between first and second year in medical school; pragmatism and fear of a federal power grab lead me in a different direction.

Sorry for the hijack.

 

[medgator]

I agree, @ my institution we would do CyberKnife (SBRT). Whenever we have previous XRT to the spine (near/at dose tolerance) SBRT can deliver a therapeutically meaningful dose while reducing the risk of complications.

You can possibly use IMRT but it would be more difficult.

Much of the spine SBRT is done using IMRT if you aren't using a cyberknife. I think that's particularly where a tomotherapy machine is well suited, when it comes to SRS/SBRT.

With things like lung and CNS, you need those non-coplanar beams to get sharper dose falloff IMO, which makes tomotherapy less suited, and a cyberknife, or an IGRT linac better suited (or a Gamma Knife if we are talking CNS).

 

[G'ville Nole]

Given the factors in this case (radiosensitive histology, intradural lesion), I'd align myself with Simul's 1st impression if he hadn't gone to surgery. After appropriate informed consent, give something like 2 Gy x 10-15. I think SBRT is reasonable for extradural lesions where you can achieve a steep dose gradient, but based on recent derivations of spinal cord a/b (all less than 2, some less than 1), I always try to keep it in the back of my mind that fractionation is your friend when it comes to the spinal cord. I'd hammer home risks vs. benefits in the consent process, but the preponderance of clinical complications data suggest that even a more aggressive fractionation scheme, along the lines of Palex's scenario (with his caveats regarding prior dose and latency noted) would be reasonable.

That, plus I'm SBRT-less for a few months in my new practice.