Low lying neck nodes and brachial plexus

[Reaganite]

Wondering what approach you guys prefer for low-lying neck nodes (i.e. those close to brachial plexus) in definitive head and neck cancer patients. At least 3 schools of thought I've seen/read about:

1. Screw you, I'm taking this node to 70.
2. I'm afraid, I'm taking this node to 60 and following with a planned neck dissection.
3. Take node to 63-66 and monitor response with PET. Offer neck dissection only for incomplete response.

---

Members don't see this ad.

 

[TarHeelRadOnc]

Wondering what approach you guys prefer for low-lying neck nodes (i.e. those close to brachial plexus) in definitive head and neck cancer patients. At least 3 schools of thought I've seen/read about:

1. Screw you, I'm taking this node to 70.
2. I'm afraid, I'm taking this node to 60 and following with a planned neck dissection.
3. Take node to 63-66 and monitor response with PET. Offer neck dissection only for incomplete response.

For HPV positive dz in a patient unlikely to need neck dissection (ie no large N2 or N3 nodes), I prefer option 3 (typically go to 66). For HPV negative pt or in pt who will likely need a neck dx anyway for bulky nodal dz, I prefer option 2 (63Gy for dose painted IMRT, 60Gy for sequential IMRT plans). I stay away from 70Gy to brachial plexus in private practice, but know that some centers do it with (reportedly) limited toxicity.

 

[ShirleyT]

I agree, I never take the plexus to 70 Gy in my practice. If bulky disease, I always go to 60 and discuss with surgeon up front that they will need neck dissection. If non bulky, I usually go to 66, watch for hot spots around the plexus and observe. Although supposodly it may be OK to take the plexus to 70, i think i saw a paper a few weeks ago in the red journal showing that brachial plexopathy was pretty rare in head and neck patients (and i think with older school larynx cancer plans with big opposed laterals, some of the plexus may have gotten 70). . I just don't feel comfortable going above 66.

 

[deleted4401]

I treat all gross disease to 70 Gy. In the era of CRT with great responses and pet-ct, I hope for the best.
I see the point of the other approaches, but brachial plexopathy is rare. I like the idea of a lower dose and planned dissection, rather than the middle approach. Who knows, though?

S

 

[medgator]

I treat all gross disease to 70 Gy. In the era of CRT with great responses and pet-ct, I hope for the best.
I see the point of the other approaches, but brachial plexopathy is rare.

Agreed

 

[RadRadRad]

question is do you think the risk of brachial plexopathy outweighs the added toxicity of neck dissection on top of RT. Certainly neck stiffness/edema/fibrosis is worse. There is also risk of anesthesia, bleeding, infection, etc.

 

[Cancerdancer]

question is do you think the risk of brachial plexopathy outweighs the added toxicity of neck dissection on top of RT. Certainly neck stiffness/edema/fibrosis is worse. There is also risk of anesthesia, bleeding, infection, etc.

The publication referenced above is PMID: 22444998. They report a 22% rate of brachial plexopathy past 5 years...and conclude it appears underreported in most literature.

Actually a tough question quoted above, as the paper indicates that neck dissection roughly doubles the rate of plexopathy, but they also provide a nice curve showing the rate of plexopathy related to dose, with the inflection point apparently ~65 Gy. A nice paper.


We typically treat to 70 Gy, and for the most part have aborted postop neck dissections. I think that's reasonable given that surgery seems to probably contribute to plexopathy (ie 60 Gy + surg ~= 70 Gy for plexopathy) while also increases operative risks and fibrosis etc...

 

[seper]

I've seen bilateral brachial plexopathy after H&N XRT and it's underreported for sure. I outline brachial plexus on all H&N patients, limit plexus Dmax to 66 Gy and keep V60 <0.5 cc.

Using IMRT and cutting PTV margin if needed, I've been able to deliver 68 Gy to gross nodes for all my patients thus far.

 

[TarHeelRadOnc]

This is a great topic and discussion. Couple additional thoughts.

1. The brachial plexus, as defined by RTOG contouring atlas, extends well above the low neck... into the mid-upper level III. I would definitely take an involved level III node to 70Gy. I would only back off dose for low neck/SCV nodes near the subclavian vessels.

2. I am convinced that 66Gy+chemo is (more than?) adequate for HPV positive disease or for small nodes (<1.5cm). Hard to prove, but I think that therapeutic ratio favors slight (6% to 66Gy) dose reduction to low neck dz in these patients to reduce BP risk

3. 66Gy prescribed to volume with 3D planning is roughly equivalent to 70Gy prescribed to isocenter... which is how radiation was prescribed in most of the trials that we use to justify our current tx recs. I'm not saying we should treat all gross dz to 66Gy... just helping rationalize a 66Gy and observe approach.

4. Most patients with large N2 (>5cm) or N3 dz need neck dx anyway. The available data using CT (UF) or PET (Iowa, MDACC, MSKCC) to determine need for neck dx did not include many patients with bulky adenopathy (and almost no N3 patients), and I would argue that neck dx (despite its increased morbidity) remains the standard in these high risk pts. As such, we can determine who is likely to need a post-tx neck dx up front. If you have a patient who is going to need an ipsilateral neck dx anyway, it makes little sense to give 70Gy to the low neck followed by neck dx... risk of BP certainly higher with 70Gy + Neck dx vs 60Gy + neck dx, right?

5. Very steep (quadratic) rise in modelled risk of BP in aforementioned UC Davis study between 66Gy and 74Gy. Max point dose to BP with prescription of 70Gy to low neck (depending on location of gross disease) will be at least 70Gy, and would likely be in the range of 72-74Gy.

6. Would be hesitant to say 70Gy to low neck in Louisville

 

[deleted4401]

Really good points and I think I'll make some modifications based on the discussion.

The problem with guessing who needs neck dissection is that it is just so hard to predict who's not going to respond. Even big N2s and some N3s might just melt away, and as is being said, the combination of surgery and RT (even 60 Gy) is worse than either modality alone (even 70 Gy).

In that paper, it says that neck dissection was an independent risk factor for BP. I wonder what the rate of BP was for patient's treated with and without neck dissection. 22% sounds high, their n must have gotten a lot smaller when they took out all the people not followed for at least 5 years. However, my n isn't all that high with 2 years of practice.

You really think 70 Gy would get you dinged in Louisiville? I think some head and neck experts would definitely do it that way and it's an acceptable variation. Maybe not.

 

[TarHeelRadOnc]

Really good points and I think I'll make some modifications based on the discussion.

The problem with guessing who needs neck dissection is that it is just so hard to predict who's not going to respond. Even big N2s and some N3s might just melt away, and as is being said, the combination of surgery and RT (even 60 Gy) is worse than either modality alone (even 70 Gy).

In that paper, it says that neck dissection was an independent risk factor for BP. I wonder what the rate of BP was for patient's treated with and without neck dissection. 22% sounds high, their n must have gotten a lot smaller when they took out all the people not followed for at least 5 years. However, my n isn't all that high with 2 years of practice.

You really think 70 Gy would get you dinged in Louisiville? I think some head and neck experts would definitely do it that way and it's an acceptable variation. Maybe not.

I think that saying 70Gy is rolling the dice. If you get a HNCa examiner who does it then your fine. If not, your saying that you would exceed tolerance (Emami and Quantec) of a critical structure when alternative options exist.

Not sure that 70Gy is safer than 60Gy + neck dx. From UCD study, which I agree is best recent study on the topic: Odds ratio of BP for neck dx is 9.55. Odds ratio of BP per Gy max dose is 1.87, or 18.7 for 10Gy increased dose.

 

[Reaganite]

This is a great topic and discussion. Couple additional thoughts.

3. 66Gy prescribed to volume with 3D planning is roughly equivalent to 70Gy prescribed to isocenter... which is how radiation was prescribed in most of the trials that we use to justify our current tx recs. I'm not saying we should treat all gross dz to 66Gy... just helping rationalize a 66Gy and observe approach.

I think this is an excellent point and one that I've used to justify giving 66Gy to PTV and observing. This is true for so many disease sites actually. I can't remember if I posted this here before, but I recently saw a meningioma that was close to the optics. A colleague insisted that we had to take the ptv (and therefore the optics) to 54Gy because there was old school data showing a dose response above 52Gy. While I think you can safely take the optics to 54 Gy, I'm not in favor of going that high if I don't have to, so I explained to her that that 54 Gy was likely to isocenter which would have meant PTV doses more around the 50Gy range. When you're right at that inflection point of these toxicity curves, these differences of a few Gy actually matter and I totally agree that it's important to remember that the old trials were isoenter rxs.

 

[Reaganite]

P.S. Always appreciate the great responses/discussions here. You guys are awesome.

 

[temujim]

The quoted paper from UCD used a patient reported survey rather than neurologic testing - their number may be an overestimate. I do a lot of H&N and I just don't hear complaints about symptoms suggestive of brachial plexopathy. Folks that I talk to that do a lot of H&N tell me the same thing. I routinely give 70 Gy. Don't even contour the brachial plexus. That structure is a huge pain.

The dose discussion is interesting, and I think it is acceptable to give nodes less than 3-4 cm 66 Gy.

 

[IndyXRT]

I see a fair amount of H&N patinets. I don't think twice about 70 Gy to a level III node. In level IV or SCV, I will try to spare the brachial plexus if possible, but let tumor coverage win. The more lateral the node, the more I worry (I do not believe the brachial plexus has a uniform dose response; nerve roots would seem to be less sensitive). I have seen way more toxicity from nodal dissection than from brachial plexopathy (plexopathy n=0 for patients I have treated n=1 for patients I picked up in follow-up), but one could argue I haven't been following my patients long enough yet.

 

[subatomicdoc]

I think that saying 70Gy is rolling the dice. If you get a HNCa examiner who does it then your fine. If not, your saying that you would exceed tolerance (Emami and Quantec) of a critical structure when alternative options exist.

Not sure that 70Gy is safer than 60Gy + neck dx. From UCD study, which I agree is best recent study on the topic: Odds ratio of BP for neck dx is 9.55. Odds ratio of BP per Gy max dose is 1.87, or 18.7 for 10Gy increased dose.

I agree. Here's the link to the UCD study abstract: http://1.usa.gov/QNoIN8 If you look at RTOG protocols, the Dmax (to 0.03 cc) ranges from 60-66 Gy depending upon the trial.

On the orals I think it shows awareness of this dilemma to say you'd work with the ENT to follow closely for the need for neck dissection, but I'd be hesitant to push above 66 Gy. I use QUANTEC as a helpful guide, too.

 

[Beamarati]

what about the use of induction TPF??? i know we still lacking enough data to such practice but we ( at KHCC) often use this approach (TPFx3) in bulky nodal disease and if we consider using the post chemo volume for the high dose volume we might have a chance to lower the brachial plexus doses and deliver 70 Gy.

 

[Gfunk6]

what about the use of induction TPF??? i know we still lacking enough data to such practice but we ( at KHCC) often use this approach (TPFx3) in bulky nodal disease and if we consider using the post chemo volume for the high dose volume we might have a chance to lower the brachial plexus doses and deliver 70 Gy.

The problem is that nobody knows what to do if TPF results in a cCR. Can you drop the dose or can you use the post-chemo treatment volumes? If not, what was the point of doing induction in the first place?

 

[temujim]

what about the use of induction TPF??? i know we still lacking enough data to such practice but we ( at KHCC) often use this approach (TPFx3) in bulky nodal disease and if we consider using the post chemo volume for the high dose volume we might have a chance to lower the brachial plexus doses and deliver 70 Gy.

Two trials presented at ASCO this year on +/- induction TPF (Paradigm, Decide) were both negative. No reason to risk the 1-4%+ mortality rate with induction TPF without any proven benefit.

 

[medgator]

Two trials presented at ASCO this year on +/- induction TPF (Paradigm, Decide) were both negative. No reason to risk the 1-4%+ mortality rate with induction TPF without any proven benefit.

Perhaps he's using the Spanish data presented at asco a few years ago?

www.asco.org/ASCOv2/Meetings/Abstracts?&vmview=abst_detail_view&confID=65&abstractID=31776

I personally would wait until some long term USA data is available

 

[Palex80]

Perhaps he's using the Spanish data presented at asco a few years ago?

www.asco.org/ASCOv2/Meetings/Abstracts?&vmview=abst_detail_view&confID=65&abstractID=31776

I personally would wait until some long term USA data is available

I see two problems with the Spanish data:

1. They have never been published as a full paper, although the abstract was a couple of years ago. Strange...

2. The "standard" RCT-arm has miserable OS results, lower, than what one would expect from daily practice.

 

[TarHeelRadOnc]

I see two problems with the Spanish data:

1. They have never been published as a full paper, although the abstract was a couple of years ago. Strange...

2. The "standard" RCT-arm has miserable OS results, lower, than what one would expect from daily practice.

Another concerning feature of the abstract from the Spanish/Hitt trial is that the data was reported "as treated" rather than "intent to treat." This kind of analysis of a randomized trial introduces significant bias. From what I understand (I wasn't there) this was the source of much criticism at the ASCO presentation and may be part of the reason the trial has yet to be published in a peer reviewed journal.

 

[RadRadRad]

agree with tarheel. pretty sure some patients seemed to disappear off the Hitt trial...ie a floating denominator. This may have been more noticeable in the oral presentation than in the abstract itself???

In regards to Decide trial, it wasn't totally negative. There was a ss reduction in distant metastases (10% vs 19%), however no difference in OS (at least with current follow up)

 

[medgator]

agree with tarheel. pretty sure some patients seemed to disappear off the Hitt trial...ie a floating denominator. This may have been more noticeable in the oral presentation than in the abstract itself???

In regards to Decide trial, it wasn't totally negative. There was a ss reduction in distant metastases (10% vs 19%), however no difference in OS (at least with current follow up)

The decide trial has that favored Chicago chemo regimen. Not sure how applicable it will be to the mainstream

 

[temujim]

agree with tarheel. pretty sure some patients seemed to disappear off the Hitt trial...ie a floating denominator. This may have been more noticeable in the oral presentation than in the abstract itself???

In regards to Decide trial, it wasn't totally negative. There was a ss reduction in distant metastases (10% vs 19%), however no difference in OS (at least with current follow up)

Though Decide did report a 10 vs 19% improvement in DM this was the subset of DM without locoregional recurrence. No ss difference in straight DMFS. While most outcomes in their trial favored the DPF arm (none signficantly though), I think few people will adopt the DFHX Chicago regimen.

The Paradigm trial, which had an arm closer to most H&N practices (72Gy CB + cisx2) favored the non-induction DPF arm in most outcomes. How these investigators can still go out and say induction DPF is still a standard of care is beyond me.

 

[Beamarati]

I know that's induction is not the standard and we Don't use it.... unless in slected cases and of course the volumes that we should be used is pre chemo volumes... I was just wondering if someone is considering boosting the post chemo volume in such case? Theoretical if there's a good response u will spare the brachial plexus.... what do u think?