Role of SABR in widely metastatic disease

[Gfunk6]

SABR-COMET looked at upfront oligometastatic disease treated with systemic therapy +/- SABR showing PFS/OS benefit. The follow-up trial will look at 6-10 mets.

How do you guys handle widely metastatic disease including scenarios like peritoneal carcinomatosis?

I am seeing a lot of patients with widely metastatic disease who receive upfront, multiagent systemic therapy. This goes on for months/years, they have a good response and eventually are left with 1-3 sites of persistent disease. We have been ablating these to improve PFS and also to get patients on chemo holidays and thereby improve QoL.

The logic is that these patients have had the "tincture of time" to show that (a) they don't harbor additional microscopic disease that mets out further and (b) don't have disease that is biologically resistant to the Med Onc's choice of systemic therapy. Therefore, they are "self-selecting" for SABR.

Would appreciate any articles or reviews that you guys find useful on the subject.

---

Members don't see this ad.

 

[Radiator20]

I am seeing a lot of patients with widely metastatic disease who receive upfront, multiagent systemic therapy. This goes on for months/years, they have a good response and eventually are left with 1-3 sites of persistent disease. We have been ablating these to improve PFS and also to get patients on chemo holidays and thereby improve QoL.

This strategy seems intuitively reasonable to me. Whether those goals will be borne out remains, I think, to be demonstrated. An alternative strategy, if they're going to continue on systemic therapy anyway, could be to wait for oligoprogression and then ablate the oligoprogressive lesion(s). Given the low level of evidence for either of these approaches, I think appropriate discussion and consent would be particularly key. And, of course, doing everything you can to minimize risk of SBRT toxicity.

 

[nkmiami]

SABR-COMET looked at upfront oligometastatic disease treated with systemic therapy +/- SABR showing PFS/OS benefit. The follow-up trial will look at 6-10 mets.

How do you guys handle widely metastatic disease including scenarios like peritoneal carcinomatosis?

I am seeing a lot of patients with widely metastatic disease who receive upfront, multiagent systemic therapy. This goes on for months/years, they have a good response and eventually are left with 1-3 sites of persistent disease. We have been ablating these to improve PFS and also to get patients on chemo holidays and thereby improve QoL.

The logic is that these patients have had the "tincture of time" to show that (a) they don't harbor additional microscopic disease that mets out further and (b) don't have disease that is biologically resistant to the Med Onc's choice of systemic therapy. Therefore, they are "self-selecting" for SABR.

Would appreciate any articles or reviews that you guys find useful on the subject.

This is how everyone I know would also handle the situation.

 

[evilbooyaa]

Remember that the Gomez trial didn't require people to be oligometastatic upfront - it required them to have 1-3 lesions after 4 cycles of systemic therapy - meaning that if you had 20 metastatic lesions but 18 of them vanished after initial systemic therapy, you were eligible.

While it is an extrapolation to put that same data to non-NSCLC, we do it infrequently.

The idea of truly consolidating (meaning treating when there are 1-3 mets seen, not at oligoprogression) isn't as common, at least at my institution, in diagnoses outside of NSCLC. What's much more common is that a patient will have 5+ 'stable' metastases, and then one or two will start growing (oligoprogression) and we're asked to SBRT the growing lesions.

I think of peritoneal carcinomatosis like maglinant pleural effusion and leptomeningeal disease. Even if there are some 'nodules' present, the entire peritoneal cavity is assumed to be affected. Both MPE and LMD are (at least historically) strict no-nos to any stereotactic treatment. All SRS trials exclude you if you have LMD, oligomet trials exclude those with MPE.

I would not feel comfortable SBRTing (at other sites of disease outside of the peritoneum) an oligomet patient with peritoneal carcinomatosis, unless they had something very aggressive in an attempt to manage it (at least IP chemo, likely HIPEC). I'm also not going to SBRT a peritoneal nodule.

 

[nkmiami]

Remember that the Gomez trial didn't require people to be oligometastatic upfront - it required them to have 1-3 lesions after 4 cycles of systemic therapy - meaning that if you had 20 metastatic lesions but 18 of them vanished after initial systemic therapy, you were eligible.

While it is an extrapolation to put that same data to non-NSCLC, we do it infrequently.

The idea of truly consolidating (meaning treating when there are 1-3 mets seen, not at oligoprogression) isn't as common, at least at my institution, in diagnoses outside of NSCLC. What's much more common is that a patient will have 5+ 'stable' metastases, and then one or two will start growing (oligoprogression) and we're asked to SBRT the growing lesions.

I think of peritoneal carcinomatosis like maglinant pleural effusion and leptomeningeal disease. Even if there are some 'nodules' present, the entire peritoneal cavity is assumed to be affected. Both MPE and LMD are (at least historically) strict no-nos to any stereotactic treatment. All SRS trials exclude you if you have LMD, oligomet trials exclude those with MPE.

I would not feel comfortable SBRTing (at other sites of disease outside of the peritoneum) an oligomet patient with peritoneal carcinomatosis, unless they had something very aggressive in an attempt to manage it (at least IP chemo, likely HIPEC). I'm also not going to SBRT a peritoneal nodule.

I assumed that carcinomatosis resolved and patient is now several years out with 1-3 sites of disease, but may have had very widespread disease at initial presentation. I have certainly done this in lung cancer multiple times in cases with presenting initial malignant effusions that resolved with immuno/chemo/targerted and they present with lets say a solitary adrenal mets 4 years later. To not treat it, seem unreasonable to me, if it can be done with almost no toxicity.

 

[evilbooyaa]

I assumed that carcinomatosis resolved and patient is now several years out with 1-3 sites of disease, but may have had very widespread disease at initial presentation. I have certainly done this in lung cancer multiple times in cases with presenting initial malignant effusions that resolved with immuno/chemo/targerted and they present with lets say a solitary adrenal mets 4 years later. To not treat it, seem unreasonable to me, if it can be done with almost no toxicity.

Completely agree with that approach. I meant more in the sense of ongoing (even just persistent, stable) peritoneal disease. If patient had peritoneal carcinomatosis at diagnosis that resolved that has not been visible on CT or clinically an issue for years then I do not think it is unreasonable.

 

[metallica81788]

So let's say a breast cancer patient who had widespread osseous mets at diagnosis and 10 years later on systemic therapy has developed an aysmptomatic peritoneal nodule? Imaging over years shows stable but still widespread osseous disease that hasn't done anything. How reasonable if no signs/sx of peritoneal carcinomatosis?

 

[scarbrtj]

If you have ever done a laparotomy, it’s a bit surprising to see how motile the guts (and its peritoneum) are. A peritoneal nodule is not always staying still (not related to the diaphragm moving, but that adds to movement too) and can have movement even more problematic than a lung lesion or a liver lesion. I’m only basing this on what I’ve seen with my own eyeballs, not in an IGRT situation. I’d worry about reproducible targetability. I’d like to see some academic types try it before I would cross that rubicon. As bad as I hate to say it might be a good case for MRI linac.

 

[Radiator20]

If you have ever done a laparotomy, it’s a bit surprising to see how motile the guts (and its peritoneum) are. A peritoneal nodule is not always staying still (not related to the diaphragm moving, but that adds to movement too) and can have movement even more problematic than a lung lesion or a liver lesion. I’m only basing this on what I’ve seen with my own eyeballs, not in an IGRT situation. I’d worry about reproducible targetability. I’d like to see some academic types try it before I would cross that rubicon. As bad as I hate to say it might be a good case for MRI linac.

I agree entirely with these concerns. Absent a fiducial or an MR linac, I would not SBRT a peritoneal target.

 

[Mandelin Rain]

Carcinomatosis of any organ would strike as a fairly strong contraindication to SBRT to that organ.

 

[metallica81788]

If you have ever done a laparotomy, it’s a bit surprising to see how motile the guts (and its peritoneum) are. A peritoneal nodule is not always staying still (not related to the diaphragm moving, but that adds to movement too) and can have movement even more problematic than a lung lesion or a liver lesion. I’m only basing this on what I’ve seen with my own eyeballs, not in an IGRT situation. I’d worry about reproducible targetability. I’d like to see some academic types try it before I would cross that rubicon. As bad as I hate to say it might be a good case for MRI linac.

I agree entirely with these concerns. Absent a fiducial or an MR linac, I would not SBRT a peritoneal target.

Carcinomatosis of any organ would strike as a fairly strong contraindication to SBRT to that organ.

I agree with all of these points, just wanted to get a referendum in the context of this thread. There are some types where I'm at that may be trying to push the limits of SABR-COMET so just checking in...

 

[nkmiami]

Had to treat a solitary peritoneal met several years ago. Pt still alive. Kind of big production. Used 10 x 5 gy. You need to use breath hold on cone beam otherwise can’t see anything. Also CT’d on sim before each treatment to get an idea of possible shifts and treated her in am and asked her not to eat breakfast. Obviously fiducial is best but was not feasible in this case.

 

[bubbachuck]

SABR-COMET looked at upfront oligometastatic disease treated with systemic therapy +/- SABR showing PFS/OS benefit.

I don't think that's correct.

Ref 1: https://www.redjournal.org/article/S0360-3016(18)31106-4/fulltext
"We enrolled patients who had a controlled primary malignancy with 1-5 metastatic lesions, all of which were amenable to SABR, with good performance status (ECOG 0-1) and life expectancy >6 months"

Ref 2: Stereotactic Ablative Radiotherapy for Comprehensive Treatment of Oligometastatic Tumors (SABR-COMET) - Full Text View - ClinicalTrials.gov
"Inclusion Criteria:
...

• Controlled primary tumor
• a. defined as: at least 3 months since original tumor treated definitively, with no progression at primary site"

Gomez trial was "upfront" metastatic (Stage IV) disease.

 

[Reaganite]

Speaking of Comet, I do insurance reviews and we are getting a lot of requests to use SBRT on patients with 4 or 5 mets. There are providers literally requesting 25 sessions of SBRT to sequentially treat 5 mets with 5 fraction SBRT! While the enrollment criteria allowed up to 5 mets, it looks like vast majority of patients had 3 or fewer. Has anyone seen the specific breakdown of patients who had 1 vs. 2 vs. 3 mets, etc.?

 

[evilbooyaa]

Speaking of Comet, I do insurance reviews and we are getting a lot of requests to use SBRT on patients with 4 or 5 mets. There are providers literally requesting 25 sessions of SBRT to sequentially treat 5 mets with 5 fraction SBRT! While the enrollment criteria allowed up to 5 mets, it looks like vast majority of patients had 3 or fewer. Has anyone seen the specific breakdown of patients who had 1 vs. 2 vs. 3 mets, etc.?

Nearly all 1-3 mets even in SABR-COMET (Gomez mandated 3 or less): Less than 10% of patient population had 4 or more mets.

 

[Chartreuse Wombat]

WRT SABR-COMET. Is anyone concerned that the difference between the arms can be completely explained by the imbalance in prostate cases? If memory serves 14 in one arm and 2 in the other. Given the better prognosis for prostate patients relative to lung, etc this imbalance can explain the results.

 

[CaesarRO]

WRT SABR-COMET. Is anyone concerned that the difference between the arms can be completely explained by the imbalance in prostate cases? If memory serves 14 in one arm and 2 in the other. Given the better prognosis for prostate patients relative to lung, etc this imbalance can explain the results.

They thought about this and addressed it in the presentation with a sensitivity analysis. Based on that, they claim they are as confident as they can be it's not the histology driving the results, but obviously more data is needed

 

[Gfunk6]

I don't think that's correct.

Ref 1: https://www.redjournal.org/article/S0360-3016(18)31106-4/fulltext
"We enrolled patients who had a controlled primary malignancy with 1-5 metastatic lesions, all of which were amenable to SABR, with good performance status (ECOG 0-1) and life expectancy >6 months"

Ref 2: Stereotactic Ablative Radiotherapy for Comprehensive Treatment of Oligometastatic Tumors (SABR-COMET) - Full Text View - ClinicalTrials.gov
"Inclusion Criteria:
...

• Controlled primary tumor
• a. defined as: at least 3 months since original tumor treated definitively, with no progression at primary site"

Gomez trial was "upfront" metastatic (Stage IV) disease.

Good catch

 

[Chartreuse Wombat]

I don't believe the MVA. There I said it. I doubt the assumptions of MVA were met. We shall see if this gets published in a high-tier journal but I remain very skeptical. If the HR changes that much according to whether you include/exclude populations then there is probably collinearity happening which calls into question the MVA. No, I am not a PhD statistician (just a troll on a website)

 

[evilbooyaa]

I don't believe the MVA. There I said it. I doubt the assumptions of MVA were met. We shall see if this gets published in a high-tier journal but I remain very skeptical. If the HR changes that much according to whether you include/exclude populations then there is probably collinearity happening which calls into question the MVA. No, I am not a PhD statistician (just a troll on a website)

While OS were different (HR 0.83 when excluding all prostate patients vs 0.60 when including all), PFS was essentially the same (HR 0.61 vs 0.58) per the slide posted above. At worst, this is similar to the initial Gomez publication showing PFS only benefit (which still affected treatment decisions). If you're a hardcore stickler for "nothing must be offered to the patient unless it is to be done for a survival benefit" then OK, feel free to proceed with the mental gymnastics necessary to discount this data.

 

[mavrik13]

I don't believe the MVA. There I said it. I doubt the assumptions of MVA were met. We shall see if this gets published in a high-tier journal but I remain very skeptical. If the HR changes that much according to whether you include/exclude populations then there is probably collinearity happening which calls into question the MVA. No, I am not a PhD statistician (just a troll on a website)

Since it is being published in the Lancet, that should give you some confidence. Randomization is a funny thing. Damned if you don't, and damned if you do and it leads to a random imbalance in the groups.

 

[Chartreuse Wombat]

Since it is being published in the Lancet, that should give you some confidence. Randomization is a funny thing. Damned if you don't, and damned if you do and it leads to a random imbalance in the groups.

Random imbalance happens in small studies like this. I am surprised that it will be published in Lancet. Do you this for a fact?

 

[RO2019]

Palma said at the plenary when he presented the data that it was going to be published in The Lancet.

Btw I thought Palma gave an amazing presentation and the data is super exciting - but he himself through his entire presentation urged caution in interpreting the data.

You don’t at ALL need to do mental gymnastics to discount or question the data. HIs entire worry was that people are going to overextend the meaning of this trial.

 

[scarbrtj]

Had to treat a solitary peritoneal met several years ago. Pt still alive. Kind of big production. Used 10 x 5 gy. You need to use breath hold on cone beam otherwise can’t see anything. Also CT’d on sim before each treatment to get an idea of possible shifts and treated her in am and asked her not to eat breakfast. Obviously fiducial is best but was not feasible in this case.

Radiation oncologists: the only doctors who instruct their patients not to eat AND not to breathe? 🙂

 

[Chartreuse Wombat]

Palma said at the plenary when he presented the data that it was going to be published in The Lancet.

Btw I thought Palma gave an amazing presentation and the data is super exciting - but he himself through his entire presentation urged caution in interpreting the data.

You don’t at ALL need to do mental gymnastics to discount or question the data. HIs entire worry was that people are going to overextend the meaning of this trial.

Huh...ASTRO in October 2018 and nothing in Lancet yet. I will wait for the paper to be published.

 

[ramsesthenice]

They thought about this and addressed it in the presentation with a sensitivity analysis. Based on that, they claim they are as confident as they can be it's not the histology driving the results, but obviously more data is neededView attachment 252161

Not really. They may have tried to address it but sensitivity and propensity analyses can’t make data something they are not. Remember, these are the same tests urologists have used to convince the world that even though every retrospective study showing better survival with surgery than RT included substantially older and sicker patients in the radiation arms, the survival difference is because surgery is a better cure. When they do matched analysis to “account for these differences” it’s still different. Because remember, the few variables they put in the analysis are the only possible variables used to decide what treatment people got. Nothing else. As long as we think about age and CDS, they are magically the same now.

Garbage in = garbage out. If your groups are different, they are different and no matter how fancy some of these statistical tests sound, there is not much you can really do about it. A number of high impact journals don’t do a good job with statistical review and let people make very bold and unsupported claims.

To be clear, I’m not talking about this presentation in particular. They did a good job being objective and keeping the data in context.

 

[Haybrant]

Just want to be sure I got this straight. Sabr comet is only for people that developed mets after their primaries were treated in the past (near or distant). If there is a guy that presented with limited prostate mets, started ADT and has limited persistent mets then what is the course of action. With STAMPEDE I know we would treat the prostatae up front in this situation now, but that was not the case before the results back in October. So how to handle this - pt had a good response to ADT but still has 3 bony mets, primary (prostate) is untreated.

 

[evilbooyaa]

Just want to be sure I got this straight. Sabr comet is only for people that developed mets after their primaries were treated in the past (near or distant). If there is a guy that presented with limited prostate mets, started ADT and has limited persistent mets then what is the course of action. With STAMPEDE I know we would treat the prostatae up front in this situation now, but that was not the case before the results back in October. So how to handle this - pt had a good response to ADT but still has 3 bony mets, primary (prostate) is untreated.

Yes, your understanding of SABR-COMET is correct. I would treat prostate per STAMPEDE and then extrapolate SABR-COMET and/or STOMP to say treating mets is good too. Despite the previous use of ADT I would treat.

 

[Haybrant]

Yes, your understanding of SABR-COMET is correct. I would treat prostate per STAMPEDE and then extrapolate SABR-COMET and/or STOMP to say treating mets is good too. Despite the previous use of ADT I would treat.

If med onc is sending you someone that has been on ADT for like 3-4 years w limited prostate mets what would you do? Tx per stampede then go after mets? Its not really stampede per se

Also, how do you tx long bone or pelvic bone mets w sbrt? Haven’t done it before

 

[evilbooyaa]

If med onc is sending you someone that has been on ADT for like 3-4 years w limited prostate mets what would you do? Tx per stampede then go after mets? Its not really stampede per se

Also, how do you tx long bone or pelvic bone mets w sbrt? Haven’t done it before

I, personally, would extrapolate STAMPEDE assuming they are still oligometastatic. To me, a patient who has maintained oligometastatic after a few years of ADT alone is somebody who has a better prognosis than an upfront oligometastatic (which is what STAMPEDE was). The tumor biology has been tested and the patient has not progressed.

I would (likely simultaneously rather than sequentially) treat by extrapolating SABR-COMET/STOMP. I don't have phase III randomized data to support these extrapolations, but I don't think in the setting of the STAMPEDE trial we're going to see additional trials that do upfront systemic therapy.

Most pelvic locations aren't an issue - 9-10Gy x 3. Long bones I somewhat worry about, but it seems people are doing 8-10 Gy (maybe more likely 8 or 9 rather than 10 Gy) x 3. Some people also do single fraction regimens, but I'm personally not a huge fan of that.