Lobectomy vs SBRT in patients age 80+

[Gfunk6]

Found this abomination of a retrospective study via Doximity.

https://www.annalsthoracicsurgery.org/article/S0003-4975(20)31448-X/fulltext

It concluded lobectomy offers greater survival relative to SBRT in patients age 80+. We know that all retrospective studies need to be taken with a grain of salt, but how can you take a publication seriously when there is a 97% (lobetomy) to 3% (SBRT) lopsided distribution?

---

Members don't see this ad.

 

[Doctorer]

Saw this as well. I looked at this a few years ago and the imbalances were just too large to warrant publishing.

Guess they didn't think so!

 

[FrostyHammer]

What else do you expect? Surgeon reviewers at surgical journals love to hear that surgery is the best thing in the world. Who cares about methodology or scientific rigor?

 

[Blastot]

"only patients with no comorbidities were selected", that's pretty impressive for patients 80+ and would apply to none of my patients 80+

 

[elementaryschooleconomics]

"only patients with no comorbidities were selected", that's pretty impressive for patients 80+ and would apply to none of my patients 80+

If I'm reading that right, they found over 9,000 total patients aged 80+ with NO comorbidities? I agree...that's unbelievably impressive.

Literally unbelievable.

 

[w00tz]

I think I read somewhere that for the database, 0 comorbidities is the same as unknown.

 

[evilbooyaa]

The fact they identified 9k patients over age 80 with zero co-morbidities is funny. But yes, w00tz is correct on 0 being same as unknown in NCDB.

What else do you expect? Surgeon reviewers at surgical journals love to hear that surgery is the best thing in the world. Who cares about methodology or scientific rigor?

Bingo. This is the real answer.

 

[elementaryschooleconomics]

The fact they identified 9k patients over age 80 with zero co-morbidities is funny. But yes, w00tz is correct on 0 being same as unknown in NCDB.



Bingo. This is the real answer.

"Refused surgery and opted for SBRT instead"

"Well Mrs X, because of your [unknown/undocumented comorbidity], we could operate but it's probably not in your best interest"

Great methodology guys, strong work.

 

[evilbooyaa]

Eh, I mean they're rehashing the same talking point but the concept of only looking at SBRT patients that surgeons WANTED to operate on and the patient said "no one is putting a scalpel in me!" is the closest they can come. Of course it's still not equivalent, but it is addressing a previous criticism of these studies (albeit not the 'most important' one)

 

[Burt Radnolds]

Nothing surprising here. If you enrich the surgical population by selecting out the Stage 2 and 3s by dissecting 7+ nodes, the patients are going to do better because they are truly Stage I. If you don't address the nodes at all, the population is no longer enriched and SBRT is just as good as surgery, and probably better than a crappy surgery like lobe with no nodal assessment or sublobar resection.

 

[scarbrtj]

We're basically just gonna argue about this sh*te forever. Yay.

SABRTooth: a randomised controlled feasibility study of stereotactic ablative radiotherapy (SABR) with surgery in patients with peripheral stage I nonsmall cell lung cancer considered to be at higher risk of complications from surgical resection - Pu

We conclude that a phase III RCT randomising higher risk patients between SABR and surgery is not feasible in the National Health Service. Patients have pre-existing treatment preferences, which was a barrier to recruitment. A significant proportion of patients randomised to the surgical group...

pubmed.ncbi.nlm.nih.gov

 

[medgator]

Are you guys not seeing enough lung sbrt in the septa/octagenarian crowd? Our surgeons are pretty reasonable here.

The only bad actors around us are the IR docs that think RFA is equivalent to surgery/SABR

 

[scarbrtj]

Are you guys not seeing enough lung sbrt in the septa/octagenarian crowd?

I think this study looked at over an about 11 year time period. In that period, 9000 septa/octogenarians got surgery or SBRT for lung ca; 8700 got surgery and 300 got SBRT.

This means 30 people in the "septa/octogenarians crowd" get SBRT in the USA per year. The NCDB captures about 70% of all cases nationwide, so let's generously bump this to 60/year.

We have 5200 radiation oncologists in America.

This means every radiation oncologist in America should have about a 1% chance of seeing an 80+ yo for SBRT each year. Or on average you'll have about one of these per every other MD's entire career, from med student to retirement. Roughly... because the number of rad oncs is growing, rad onc MDs may (or likely will) go whole careers and never see an 80+ yo for SBRT! "In the very distant future, the universe will have expanded so much that the light from all of the stars and galaxies outside of our galaxy group will never reach earth..." and this is why expansion is bad kids.

 

[BobbyHeenan]

Are you guys not seeing enough lung sbrt in the septa/octagenarian crowd? Our surgeons are pretty reasonable here.

The only bad actors around us are the IR docs that think RFA is equivalent to surgery/SABR

It’s like you’re there at my tumor board. The IR lung cancer RFA I thought was like a snipe but it exists, it’s real, and it’s not spectacular.

 

[medgator]

I think this study looked at over an about 11 year time period. In that period, 9000 septa/octogenarians got surgery or SBRT for lung ca; 8700 got surgery and 300 got SBRT.

This means 30 people in the "septa/octogenarians crowd" get SBRT in the USA per year. The NCDB captures about 70% of all cases nationwide, so let's generously bump this to 60/year.

We have 5200 radiation oncologists in America.

This means every radiation oncologist in America should have about a 1% chance of seeing an 80+ yo for SBRT each year. Or on average you'll have about one of these per every other MD's entire career, from med student to retirement. Roughly... because the number of rad oncs is growing, rad onc MDs may (or likely will) go whole careers and never see an 80+ yo for SBRT! "In the very distant future, the universe will have expanded so much that the light from all of the stars and galaxies outside of our galaxy group will never reach earth..." and this is why expansion is bad kids.

I treat 1-2 lung SBRT patients a month... Can't say i remembered to check their ages though

 

[scarbrtj]

I treat 1-2 lung SBRT patients a month... Can't say i remembered to check their ages though

I think there's about 15K Stage I lung cancers per year, which works out to about 3 patients per rad onc per year. So at 18 per year you're definitely WALLOWING in Stage I lung SBRT lol. I'm probably more like 4-8/year pure Stage I lung... I'm prob double that in "lung SBRT" what with the met zapping and all.

 

[OTN]

It’s like you’re there at my tumor board. The IR lung cancer RFA I thought was like a snipe but it exists, it’s real, and it’s not spectacular.

The equipment companies got to the pulmonary guys around here. When I talk about the data strongly favoring surgery or SBRT, "oh, of course you're going to say that, you're a radonc"

 

[evilbooyaa]

I think this study looked at over an about 11 year time period. In that period, 9000 septa/octogenarians got surgery or SBRT for lung ca; 8700 got surgery and 300 got SBRT.

This means 30 people in the "septa/octogenarians crowd" get SBRT in the USA per year. The NCDB captures about 70% of all cases nationwide, so let's generously bump this to 60/year.

We have 5200 radiation oncologists in America.

This means every radiation oncologist in America should have about a 1% chance of seeing an 80+ yo for SBRT each year. Or on average you'll have about one of these per every other MD's entire career, from med student to retirement. Roughly... because the number of rad oncs is growing, rad onc MDs may (or likely will) go whole careers and never see an 80+ yo for SBRT! "In the very distant future, the universe will have expanded so much that the light from all of the stars and galaxies outside of our galaxy group will never reach earth..." and this is why expansion is bad kids.

This is only 300 patients over the age of 80 who refused surgery. There are multitudes more that are not candidates for surgery.

But yes, truly stage I NSCLC is likely a relatively small portion of even the lung SBRT most of us are doing, as I believe we've discussed before. I probably do more oligomet lung than definitive lung SBRT. At least similar in terms of patient numbers.

 

[KHE88]

Since we are talking about stage I NSCLC, I have a clinical question for once.

When do you all feel comfortable omitting mediastinal lymph node staging beyond stage IA (which does not need lymph node eval per guidelines)?

I have an 85 yo with a 1.2 cm squam in the middle of the lung. He has every cardiac and pulmonary comorbidity in the book and is happily putting down at least a pack a day. He was sent to surgery for mediastinoscopy and potential surgery but had a terrible GI bleed and they finally realized that was a terrible idea to begin with and sent him to me. Never had nodes evaluated and the tiny lung tumor is the only thing active on PET.

It seems reasonable to not waste anymore time and risk complications sampling nodes on this guy.

So the question is, when do you look at nodes, and if so how do you do it (EBUS vs. mediastinoscopy vs. PET alone). This is particularly challenging in a rural setting where pulm and thoracic surgery talent and timeliness is sketch at best.

 

[evilbooyaa]

In the case described above I would leave him alone and treat what you see. Rate of occult nodal disease in 7th edition AJCC T1N0 patients was 10% in those with a negative PET/CT (Risk factors for occult nodal metastasis in clinical T1N0 lung cancer: a negative impact on survival), and for somebody with T1b disease, risk is probably slightly lower, as they noted that increasing size of tumor lead to increased risk of occult nodal involvement.

I generally favor EBUS (never mediastinoscopy unless there's an prevascular or AP window one that EBUS won't be able to reach), but will treat PET only after discussion with patient regarding a 5-10% risk of occult nodal (which may be incompletely sampled by something like an EBUS anyway).

 

[communitydoc13]

Since we are talking about stage I NSCLC, I have a clinical question for once.

When do you all feel comfortable omitting mediastinal lymph node staging beyond stage IA (which does not need lymph node eval per guidelines)?

I have an 85 yo with a 1.2 cm squam in the middle of the lung. He has every cardiac and pulmonary comorbidity in the book and is happily putting down at least a pack a day. He was sent to surgery for mediastinoscopy and potential surgery but had a terrible GI bleed and they finally realized that was a terrible idea to begin with and sent him to me. Never had nodes evaluated and the tiny lung tumor is the only thing active on PET.

It seems reasonable to not waste anymore time and risk complications sampling nodes on this guy.

So the question is, when do you look at nodes, and if so how do you do it (EBUS vs. mediastinoscopy vs. PET alone). This is particularly challenging in a rural setting where pulm and thoracic surgery talent and timeliness is sketch at best.

I would treat what you see and am always fine with a negative mediastinum on PET for older/sick pts. Negative predictive value of PET in this setting is very good. The positive predictive value is much more shaky and I prefer pathologic confirmation of PET positive nodes in mediastinum or hilum unless they are radiographically very pathologic appearing. (False positive hilar areas happen frequently).

There is data out of Italy at high volume EBUS centers that indicates that the added value of EBUS in a negative PET is not very much. And this data from MDACC indicated EBUS not that great at finding N2 disease in PET N0/N1 disease. Diagnostic performance of endobronchial ultrasound-guided mediastinal lymph node sampling in early stage non-small cell lung cancer: A prospective study - PubMed

The number one cause of death for resected early stage non-small cell lung cancer remains non-small cell lung cancer. These are surgically staged patients. I don't think there is much value to being extra vigilant with mediastinal staging in the 80+ year olds that we see.

 

[FrostyHammer]

Since we are talking about stage I NSCLC, I have a clinical question for once.

When do you all feel comfortable omitting mediastinal lymph node staging beyond stage IA (which does not need lymph node eval per guidelines)?

I have an 85 yo with a 1.2 cm squam in the middle of the lung. He has every cardiac and pulmonary comorbidity in the book and is happily putting down at least a pack a day. He was sent to surgery for mediastinoscopy and potential surgery but had a terrible GI bleed and they finally realized that was a terrible idea to begin with and sent him to me. Never had nodes evaluated and the tiny lung tumor is the only thing active on PET.

It seems reasonable to not waste anymore time and risk complications sampling nodes on this guy.

So the question is, when do you look at nodes, and if so how do you do it (EBUS vs. mediastinoscopy vs. PET alone). This is particularly challenging in a rural setting where pulm and thoracic surgery talent and timeliness is sketch at best.

There are plenty of (retrospective) data corroborating that PET staging alone vs PET+EBUS doesn't change any outcomes. In the rural setting where there are too many diagnostic-related delays and starting patients on time is very important, I'm very comfortable with PET staging alone for nearly all patients (perhaps the exception would be larger, 4+ cm tumors but that's not an exact science).

 

[metallica81788]

There are plenty of (retrospective) data corroborating that PET staging alone vs PET+EBUS doesn't change any outcomes. In the rural setting where there are too many diagnostic-related delays and starting patients on time is very important, I'm very comfortable with PET staging alone for nearly all patients (perhaps the exception would be larger, 4+ cm tumors but that's not an exact science).

Great points. Also consider rural MDs newly adopting diagnostic EBUS/bronch with low yields to where you're not even really sure if you believe the results anyway... Might as well just go with PET in that case.

As you mentioned - in the real world some extended workups for pure academic purpose only harm the patient and delay treatment.

 

[scarbrtj]

As you mentioned - in the real world some extended workups for pure academic purpose only harm the patient and delay treatment.

"You go to war with the army you have, not the army you want." - Don Rumsfeld

 

[Palex80]

Our surgeons still operate this patients and their results are not that bad. Altough SBRT is clearly very comfortable let us not forget that the thoracic surgeons have also made advances. VATS-lobectomy when done right means that the patient can go home within a few days, have minimal pain and so on...

 

[KHE88]

There are plenty of (retrospective) data corroborating that PET staging alone vs PET+EBUS doesn't change any outcomes. In the rural setting where there are too many diagnostic-related delays and starting patients on time is very important, I'm very comfortable with PET staging alone for nearly all patients (perhaps the exception would be larger, 4+ cm tumors but that's not an exact science).

Great points. Also consider rural MDs newly adopting diagnostic EBUS/bronch with low yields to where you're not even really sure if you believe the results anyway... Might as well just go with PET in that case.

As you mentioned - in the real world some extended workups for pure academic purpose only harm the patient and delay treatment.

Thanks! This is helpful to confirm I'm not crazy.
Yes, that's something that they don't teach you regarding timeliness of treating the patient vs. trying to make sure workup is done by the book. When it can take over a month to get a PET scan because you work for a garbage hospital, then it does tend to change things. Unfortunately a lot of what I see is med onc dealing with this by going rogue and giving inappropriate systemic treatment or bypassing a surgical referral when the patient really needs it (I can't tell you how many times I get sent a symptomatic 3+ cm brain met and am told to SRS it).