Dutch trial for preop CRT for esophageal and GEJ

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
This appears to support the results of the Dublin and CALGB trials - use pre-op CRT for resectable esophageal adenocarcinomas. However, the dose in this trial was 10% less than the other two (41.4 Gy vs 50.4 Gy).
 
Members don't see this ad :)
This is already making an impact on practice, as some RadOnc's are now stopping at 41.4 Gy when intent is preop and carbo/taxol is given. I disagree and still take thee pts to 50.4 Gy.
 
pCR in Dutch was 29% (41.4 Gy). pCR in CALGB was 40% (10/25 patients) (50.4 Gy). pCR in Irish trial was under 25% (40.5 Gy). So, not that this is truly scientific, but I think I'd still go higher. Been splitting difference and going to 45ish.
 
seper said:
This is already making an impact on practice, as some RadOnc's are now stopping at 41.4 Gy when intent is preop and carbo/taxol is given. I disagree and still take thee pts to 50.4 Gy.
Click to expand...

Yeah, I think there is a lot more to the dose issue than meets the eye. On one hand, it seems easy enough to say that we shouldn't adopt the 41.4 Gy because it isn't a high enough dose...particularly if a patient ends up not being a candidate for surgery. On the other hand, there are some who feel the survival advantage in this trial may have related to the very fact that a lower radiation dose was used. Treating to 41.4 means you can pretty much treat AP/PA the whole way and thus the dose to the lung, which seems to correlate with periop mortality per some data out of MDACC, is reduced.
 
Interesting thought, but i don't think that's the reason it showed a benefit. One could make the argument that multifield technique delivered throughout lowers the fractional dose of the exposed normal lung (i.e. the reason we do 4 fields at a time for pelvis, instead of 2 fields one day, 2 fields the next), and b/c late toxicity correlates with fractional dose, 2 field might not be best idea.

Of the 6 trials in the modern era, there are now 3 trials that show a benefit. the TROG, urba and bossett used non-standard RT schedules, but none showed a detriment. I think, basically, preop CRT works, if done properly, and doses of 41.4 to 50.4 represent an acceptable norm.

I think more interesting question is figuring out who needs surgery after CRT, i.e. identifying complete responders without surgery.
 
SimulD said:
Interesting thought, but i don't think that's the reason it showed a benefit. One could make the argument that multifield technique delivered throughout lowers the fractional dose of the exposed normal lung (i.e. the reason we do 4 fields at a time for pelvis, instead of 2 fields one day, 2 fields the next), and b/c late toxicity correlates with fractional dose, 2 field might not be best idea.

Of the 6 trials in the modern era, there are now 3 trials that show a benefit. the TROG, urba and bossett used non-standard RT schedules, but none showed a detriment. I think, basically, preop CRT works, if done properly, and doses of 41.4 to 50.4 represent an acceptable norm.

I think more interesting question is figuring out who needs surgery after CRT, i.e. identifying complete responders without surgery.
Click to expand...

In the MDACC data, it was actually the V5 Lung that was the important parameter. So 4 field may lower the high dose volume, but this would be at the expense of V5. I'll have to dig up the paper, but it was referenced in one of the Astro Refreshers. Not sure how much an additional 10 Gy or so off cord is gonna make to the V5, but one of the discussants at last year's ASTRO meeting brought this up.
 
Last edited:
Lower dose makes some sense when plan is trimodality. Particularly in GEJ as dose to stomach is a consideration with gastric pull-through anastamosis. Afterall, anastamotic leak is a big contributor to periop mortality and this could likely be decreased with lower dose.

I'm a believer in RT but surgery is also pretty good local treatment. The question we all must ask is what is the added benefit of those 5 fractions vs potential harm. This study has impressive results with only 4% postop mortality while prior studies with higher dose has been closer to 9-10%. Probably not all due to the lower dose but perhaps cautious respect of dose limits of the stomach may limit our perioperative m/m. Our standard thresholds may be too lenient in the face of manipulation of tissue under tension..
 
Last edited:
We usually give 45 Gy in 1.8 Gy/d for neoadjuvant treatment of esophageal cancer.
Typical chemotherapy is Cisplatin/5FU, but we are currently treating patients in study protocols with a Cisplatin/Docetaxel combination too.

Typical field arrangement if 4-field-box with at least 80% of the dose coming in ap/pa.

Two points:
1. Postoperative mortality is a big issue in these patients. One has the impression that the reported mortality rates are higher in multi-center trials, perhaps because the quality of surgery and post-surgery intensive care may be suboptimal in low-volume centers.
2. If the patient does not undergo operation after neoadjuvant treatment, because the tumor is deemed unresectable, then you can always give another round of around 25 Gy with 2 cycles of cisplatin after the break. I've seen this happen twice in my career and both patients faired miserably. Many people will tell you this is a very bad approach, because of the break in treatment and the additional toxicity to the patient (sometimes the tumor is deemed unresectable after the patient is opened up). However, one has to bear in mind, that a patient who showed little tumor regression after 45 Gy would probably not have faired a lot better if he continued the radiochemotherapy as a definitive regime. Another 10 Gy are not going to save the patient, if the tumor did not shrink after the first 45 Gy.
One could even provocatively speculate that these patients, whose tumors actually shrink the less are the ideal candidates for surgery, since surgery is the only therapy, which may heal these patients.
 
Lung dose in a preop patient is a perfectly valid concern.

Here is my issue with 41.4 Gy. In this trial, 94% pts in the CRT arm had surgery. In my practice (a large academic hospital), proportion of pts labeled as "preop" ultimately having esophagectomy is lower (? 60-80 % probably). So if I give 41.4 Gy, there is high risk that it's ultimately undertreatment and deviation from standard practice.
 
Last edited:
Members don't see this ad :)
seper said:
Lung dose in a preop patient is a perfectly valid concern.
Here is my issue with 41.4 Gy. In this trial, 94% pts in the CRT arm had surgery. In my practice (a large academic hospital), proportion of pts labeled as "preop" ultimately having esaphagectomy is lower (? 60-80 % probably). So if I give 41.4 Gy, there is high risk that it's ultimately undertreatment and deviation from standard practice.
Click to expand...

Dear Seper,
first of all I think you need to look into why so many patients labelled as "preop" in your institution end up being irresectable and what you can do to change that.
Having worked in probably one of the biggest esophageal cancer centers in Europe (probably 5 resections/week), I've found that these are the necessary steps that may be helpful:

1. The implementation of scores to predict postoperative mortality and thus filter out these patients the surgeons will back off from operating may be a good idea, for example with something like this: http://www.ncbi.nlm.nih.gov/pubmed/9667720.

2. Communicating to your surgical colleagues that the resection of the tumor "no matter what" is not the ultimate goal, when dealing with esophageal cancer, may be helpful. We have 2 randomized studies showing that definitive RCT is equivalent to neoadjuvant RCT followed by resection.
http://www.ncbi.nlm.nih.gov/pubmed/15800321
http://www.ncbi.nlm.nih.gov/pubmed/17401004
I know that's a tough task, but it has to be done.

3. For candidates you think may not be the optimal candidates for resection, you may choose to try another approach: Have an early evaluation during treatment with (PET-)CT +/-endoscopy and an assessment by the anesthesiologists / cardiologists / pulmonologists to examine tumor response to treatment and deem if the patient is functionally able to undergo resection at all.
http://www.ncbi.nlm.nih.gov/pubmed/21233607

4. Talk with your patients about alternatives. I have found out, that some patients only want to undergo resection, because they were told, that was the only way to heal themselves from the tumor. Such patients may opt not to undergo surgery, especially if the tumor is responding to RCT. This may however backfire on you, some surgeons will state you are stealing patients from them.
 
It's not easy to tell a big-shot thoracic surgeon to back off after initial consult! It's much easier to treat pt to 50.4 Gy - definitive CRT dose and until recently standard preop dose - and let Surgery decide what's next.
 
seper said:
It's not easy to tell a big-shot thoracic surgeon to back off after initial consult! It's much easier to treat pt to 50.4 Gy - definitive CRT dose and until recently standard preop dose - and let Surgery decide what's next.
Click to expand...

Like I said, "that's a tough task".

However:

1. 50.4 Gy as preoperative dose is a bit too high IMHO. Most trials used lower doses and we are not certain if higher doses could result to higher postoperative complication rates.

2. 50.4 Gy is IMHO a bit too low as for definitive treatment. I know the Minsky trial was negative for dose escalation, but this trial and its interpretation have well known methological handicaps. We usually give 54 Gy and in some cases go higher up to 59.4 Gy or give a brachy-boost. Evidence for this approach comes from several Phase-II trials.
Bear in mind that the randomized trials showing that definitive CRT is equivalent to CRT+surgery (Stahl and Bedenne), used a dose higher than 50.4 Gy to achieve this goal in the definitive setting.
The question "What is the optimal dose for definitive CRT in esophageal cancer?" has not been answered properly so far. It's a bit like the question "What is the optimal dose for °II-astrocytoma RT?" We have two randomized trials showing that 45 Gy and 50.4 Gy are as good in tumor control as 59.4 Gy and 64.8 Gy respectively and combined with less toxicity, yet however most of the people rather still give 54 Gy. Or is anyone here, who gives 45 Gy for °II-astrocytoma?
 
I see the point of peri-operative mortality with higher doses, but where is this coming from? Anecdote or single-institutional series? Or did one of the major trials show that? CALGB had only one post-operative death (in surgery alone arm) and none in tri-modality. I don't see a lot of these cases in practice and in residency they got preop chemo or surgery alone. Their reported mortality rates were very low.

I think it's sort of quibbling. I don't think anyone will compare 41.4 to 50.4. Just keep that V20 low low low, and the V5 low low low and anywhere in that range is going to be considered SOC.

As far as definitive, I asked Minsky himself. Goes 54 to 59.4 with chemo. Our little practice, we've stuck to 50.4 knowing that it will be hard to defend to a jury of your peers why you went against NCCN guidelines when there was a negative trial showing no benefit (regardless of the flaws!).

S
 
Last edited by a moderator:
SimulD said:
As far as definitive, I asked Minsky himself. Goes 54 to 59.4 with chemo. Our little practice, we've stuck to 50.4 knowing that it will be hard to defend to a jury of your peers why you went against NCCN guidelines when there was a negative trial showing no benefit (regardless of the flaws!).

S
Click to expand...

Bingo. NCCN pretty clear on this. Lots of rad oncs go higher definitively, despite that.
 
Post-op mortality:

9.3% vs 0.8% in Bedenne
12% vs 4% in Bossett
10% vs 3.8% in Stahl
Higher with higher dose in Minsky (flawed)

You can make counter arguments for each of thse studies, but I do see an overall trend of increased risk with trimodality. If we have phase III data to support a lower dose, why on earth not embrace that in favor of potentially improving lung toxicity and lessening anastamotic leak risk? I guess I don't see the big up side to an additional 10 Gy. Our goal should be to get them to surgery in these cases and if 41.4 will do that, I'm a fan.

Simul even if you respect lung constraints, there's no way to respectvthe stomach if you're dealing with GEJ. That's where I personally see the possible benefit of lower dose and hopefully lower leaks which is the postop risk we worry most about. We don't see it..but we often don't "see" the complications as they are under the care of another service by then.
 
I think you're looking at the wrong variable. In CROSS/CALGB/Irish Trial and the other 3 mentioned, it was S +/- CRT. In all 3 of the positive trials, there was no difference in post-operative mortality in trimodality vs surgery alone. In the negative trials, no difference in post-op complications/mortality with Urba or Australian trial. Worse in Bosset, but 37 Gy in 10 Fx - ugh... The other trials you mentioned were not comparison of trimodality vs surgery alone.

- Bedenne randomized between CRT +/- Surgery, not +/- CRT. I agree, adding surgery will be more toxic, that's why I'm saying it would be nice to know who doesn't need surgery.

- Stahl randomized between Chemo then CRT +/- Surgery. I agree, adding surgery will be more toxic, that's why I'm saying it would be nice to know who doesn't need surgery.

- Minsky was comparing bimodality therapies. Majority of deaths before reaching even 50.4 Gy in the experimental arm.

Two things we know: 1) In all positive trials comparing S vs S+CRT, there was no worse post-op outcome with trimodality treatment. 2) those that have pCR do better and may not need surgical intervention. If there is a way to achieve that safely and with some certainty (which we don't have, I guess they are studying PET-CT and biopsies for this), could be beneficial. I just don't see how the results of CROSS in any way contradict the results of CALGB. They both say the same thing: in selected patients, trimodality better than surgery alone. Maybe would have been even better results with higher dose in CROSS (unlikely). Again, comparing apples and oranges, the trimodality arm in CALGB has slightly better results (MS: 4.5 years vs 4.1 years, and higher pCR: 40 vs 32%). I'm not saying one approach is better or worse. They're just different.
 
In US, any treatment related morbidity that occurs when treating esophageal Ca to >50.4Gy can put you at risk of malpractice claim (sadly).
 
seper said:
In US, any treatment related morbidity that occurs when treating esophageal Ca to >50.4Gy can put you at risk of malpractice claim (sadly).
Click to expand...

Based on real world info? I wouldn't be surprised considering the NCCN recommendations and the findings from the minsky trial.
 
Yes, real cases. Also notice how at each ASTRO, after esophageal educational session, there is a question on dose escalation but speakers continue to stand firm by 50.4 Gy.
 
You must log in or register to reply here.
Next unread thread

Similar threads

Palex80
Photons vs. Protons for Esophageal Cancer
Replies
48
Views
2K
SneakyBooger
SneakyBooger
B
FLAME TRIAL prostate boost - ?lymph nodes?
2
Replies
57
Views
3K
SneakyBooger
SneakyBooger
B
NRG BR008 another day and another radiation omission trial
Replies
29
Views
2K
grenz
G
C
Yippee another Spacer trial
2 3 4
Replies
197
Views
9K
temujim
T
Palex80
Case: Esophageal Cancer with weird nodes
Replies
17
Views
920
medgator
M
Top