Locally-Advanced Duodenal Cancers

[ramsesthenice]

I know these are rare, but curious how other people are practicing. Just saw a guy who presented with a proximal tumor and multiple positive nodes. They gave him upfront FOLFOX and he actually progressed clinically (no radiographic response but worsening clinical symptoms though not even close to obstructed). I was honestly surprised how "controversial" this case was at our tumor board and I am curious what everyone would recommend?

I (successfully) argued for neoadjuvant chemorads. Some of our surgeons felt there is no role for chemorads in duodenal cancers since there is no survival benefit compared to chemo. First of all, Im not sure how relevant that argument is in someone who didn't respond to chemo. Second, there is a fairly consistent improvement in local control with chemorads (and having so many nodes his local recurrence risk is really high) in single-institution series. Finally, if you are going to be treating bowel preop intuitively makes more sense to me as I would rather treat bowel that is going to come out surgically than be left in place.

The fact this patient didn't respond well to chemo makes this case a little easier in my mind. Assuming they have the more typical PR, what would you recommend, preop CRT or surgery and observation (+/- adjuvant RT based on pathology)?

---

Members don't see this ad.

 

[PhotonBomb]

I know these are rare, but curious how other people are practicing. Just saw a guy who presented with a proximal tumor and multiple positive nodes. They gave him upfront FOLFOX and he actually progressed clinically (no radiographic response but worsening clinical symptoms though not even close to obstructed). I was honestly surprised how "controversial" this case was at our tumor board and I am curious what everyone would recommend?

I (successfully) argued for neoadjuvant chemorads. Some of our surgeons felt there is no role for chemorads in duodenal cancers since there is no survival benefit compared to chemo. First of all, Im not sure how relevant that argument is in someone who didn't respond to chemo. Second, there is a fairly consistent improvement in local control with chemorads (and having so many nodes his local recurrence risk is really high) in single-institution series. Finally, if you are going to be treating bowel preop intuitively makes more sense to me as I would rather treat bowel that is going to come out surgically than be left in place.

The fact this patient didn't respond well to chemo makes this case a little easier in my mind. Assuming they have the more typical PR, what would you recommend, preop CRT or surgery and observation (+/- adjuvant RT based on pathology)?

my only argument would be that the nodal drainage is probably closer to colon cancer (wide-spread) than rectal, so less bang for buck with radiation in terms of sterilizing nodal disease, but if your argument is for radiation to convert someone who is unresectable to resectable or to reduce chance of a positive margin, then I can see that point. however- is that the case for him? is there a reason he can't just go to surgery now?

 

[seper]

I've only seen 2 or 3 cases, but always felt surgery is central to care. Preoperative treatments may compromise surgery.

 

[thecarbonionangle]

I've only seen 2 or 3 cases, but always felt surgery is central to care. Preoperative treatments may compromise surgery.

would probably depend on timing of surgery in relation to XRT, as a general principle (preop rectal, gastric, GYN, thoracic, etc)

 

[medgator]

If you're going to radiate, will be tougher to give post op i imagine

 

[FrostyHammer]

I'm sure there's some junk NCDB study out there where the authors make it seem like their data is just a good as a hypothetical randomized trial...

 

[thecarbonionangle]

I'm sure there's some junk NCDB study out there where the authors make it seem like their data is just a good as a hypothetical randomized trial...

Hey some people make a career out of these studies. Don't knock them!

 

[ramsesthenice]

my only argument would be that the nodal drainage is probably closer to colon cancer (wide-spread) than rectal, so less bang for buck with radiation in terms of sterilizing nodal disease, but if your argument is for radiation to convert someone who is unresectable to resectable or to reduce chance of a positive margin, then I can see that point. however- is that the case for him? is there a reason he can't just go to surgery now?

I think it depends on location. For proximal tumors near the ampulla Im not sure that is true. In this case the involved nodes are in the porta and RP space so its behaving like a periampullary tumor.

And to your second question, no. Upfront surgery is absolutely an option. But the med oncs are not planning any adjuvant chemo since he didn't respond to FOLFOX. If you give upfront surgery your options will be observation (not crazy, but the risk of local recurrence is high) or post-op RT.

I've only seen 2 or 3 cases, but always felt surgery is central to care. Preoperative treatments may compromise surgery.

Agree surgery is central and this was the surgeons argument. They want to do a Whipple (which I agree with, this thing is really proximal). Based on the fact that we have extensive experience using preop for borderline pancreatic and ampullary primaries pre-Whipple, I think this particular concern is largely hypothetical but I can understand why surgery would feel this way.

 

[Neuronix]

I need to put this into a box. If it's duodenal mucosa, aka colon cancer, cut it out like a colon cancer if they think they can get negative margins and do a good lymph node dissection. If it's periampullary, I put it into a biliary thought box, and we're just back in the same argument about whether there's any role for radiation in pancreas/periampullary tumors and when to do it (neoadjuvant, adjuvant, etc). I assume since they got FOLFOX we're putting this in the mucosal thought box because FOLFOX really isn't good chemo for biliary tract tumors (I'd recommend treating more like a pancreas with gem+abraxane or adding the irinotecan--though hell I'll page @gutonc for funzies).

You have to think about large fields and doses like a post-op pancreas field in such a case in my opinion. You can do the same fields and doses pre-op, and it isn't a bad idea if you're concerned about margins up front.

Alternative is to get fancy with SBRT/short course if you're convinced it's coming out--I'd probably think about 6 Gy x 5 to gross disease and nodes, but I'd be doing it with MRI guidance.

Hey some people make a career out of these studies. Don't knock them!

Don't hate the player; hate the game.

 

[ramsesthenice]

I need to put this into a box. If it's duodenal mucosa, aka colon cancer, cut it out like a colon cancer if they think they can get negative margins and do a good lymph node dissection. If it's periampullary, I put it into a biliary thought box, and we're just back in the same argument about whether there's any role for radiation in pancreas/periampullary tumors and when to do it (neoadjuvant, adjuvant, etc).

I think that is a great way of describing the situation. Endoscopically this thing apparently looks like it started as a mucosal tumor but...
1) It will take a Whipple to get negative margins
2) Nodal drainage is to Porta and RP space as opposed to mesentary (based on grossly positive nodes)
3) It didn't respond to FOLFOX

This thing seems to be telling us we may have put it in the wrong bucket.

 

[PhotonBomb]

not responding to FOLFOX is bad for any GI tumor

 

[RickyScott]

I think that is a great way of describing the situation. Endoscopically this thing apparently looks like it started as a mucosal tumor but...
1) It will take a Whipple to get negative margins
2) Nodal drainage is to Porta and RP space as opposed to mesentary (based on grossly positive nodes)
3) It didn't respond to FOLFOX

This thing seems to be telling us we may have put it in the wrong bucket.

I agree and would give 5040 w/xeloda to large field

 

[PhotonBomb]

I agree and would give 5040 w/xeloda to large field

Why large? I would give more to less probably

 

[deleted774703]

Great case to learn from! Thanks for sharing.

My only hesitation would be RE getting potentially burned by surgeons after RT. Don't want to be on the hook for toxicity if no surgery.

 

[ramsesthenice]

I agree and would give 5040 w/xeloda to large field

I am planning on stopping at 45 to minimize risk of ulceration or bleeding. 50.4 would be fine but by the time you incorporate hot spots the risk would probably be a little higher.

Why large? I would give more to less probably

Im inclined to agree. Its acting ampullary so I was thinking about using a typical ampullary field (without a boost).

 

[Palex80]

Some of our surgeons felt there is no role for chemorads in duodenal cancers since there is no survival benefit compared to chemo.

Which is true. There is no data pointing at a possible benefit from RCT vs. CT in this disease. In principal, you can extrapolate from pancreatic cancer, where it's the same deal.

First of all, Im not sure how relevant that argument is in someone who didn't respond to chemo.

I guess, it depends on what chemo he had...
He had FOLFOX which is quite strong. Perhaps FOLFIRINOX would have been a better choice (although no data for that in this disease too and FOLFIRINOX often highly toxic). My feeling is that someone who did not respond to FOLFOX with a disease like that is actually a palliative case. You can go and perform a heroic attempt of neoadjuvant RCT and resect later, but the prognosis is probably abysmal. I would not be surprised if he progressed with distant mets during RCT, since you are probably going to give him only Capecitabine (which he already progressed on) or Gemcitabine (?).

Second, there is a fairly consistent improvement in local control with chemorads (and having so many nodes his local recurrence risk is really high) in single-institution series.

The same could be said for pancreatic cancer too. Yet, neoadjuvant RCT is not better than neoadjuvant CT for pancreatic cancer.

Finally, if you are going to be treating bowel preop intuitively makes more sense to me as I would rather treat bowel that is going to come out surgically than be left in place.

You were considering adjuvant RT for this patient? No clear data for benefit in that scenario too.

The fact this patient didn't respond well to chemo makes this case a little easier in my mind.

Indeed, it's palliative (in my view).

Assuming they have the more typical PR, what would you recommend, preop CRT or surgery and observation (+/- adjuvant RT based on pathology)?

I'd recommend surgery now (if feasible, among others as a good palliative measure, since sooner or later obstruction will become an issue) or palliative chemotherapy, that's it. I do not believe in any neoadjuvant apprach at this timepoint.
If surgery turns up into a well localized R1/R2, you could perhaps evaluate some sort of postoperative RT, if the patient recovers well. But even that approach is not well supported by clear data. And if surgery (even debulding surgery) is not possible now in dealing with obstruction (or pain, which is another common complication of tumor in that region), you can evaluate palliative RT.

 

[FrostyHammer]

If this is all being extrapolated from pancreas, it may be worth taking a look at the PREOPANC trial results, as your dilemma seems to be similar to the question asked in that trial.