Grade III Gliomas

[Palex80]

---

Members do not see this ad.

What is the current management in your clinic when it comes to treatment of Grade III Gliomas?

From what we know patients anaplastic oligodendroglioma or oligoastrocytoma, would probably benefit the most from chemotherapy (especially when they have MGMT+ or LOH1p/19q).

But what about patients with anaplastic astrocytoma without any oligo-components?

Would you recommend concurrent radiochemotherapy in all anaplastic astrocytoma patients?
Would you rather give it to the oligo-component patients? Would you decide according to MGMT or LOH1p/19q status?
Would you always give it in primary radiotherapy?
And what would you do in the postoperative setting?

 

[Gfunk6]

I think most clinicians in the US simply extrapolate the results of the Stupp trial. Interestingly, I believe the Stupp trial actually excluded Grade III. Overkill? Perhaps . . . but given the relatively low incidence of Grade III's and the historical precedent for lumping them with GBMs, I think it is appropriate.

To go even further I've seen the Stupp regimen used in Grade II's with a sizable percentage of anaplastic cells on path.

 

[deleted4401]

Eh... Why not?

Vitamin T is not very toxic, and there is a clear benefit for GBMs, so I can't see why not extrapolate for adjuvant treatment of AA, regardless of oligo component or MGMT/1p19q status. Even for lower grade, no data, but nothing really works so if you're for some reason treating with RT post-operatively (large tumor, crossing midline, recurrence, whatever), why not just add it? I can't see it hurting and nothing else really helps.

The thing that always makes me a little concerned about that Stupp trial, although it wasn't stratified for this and wasn't a planned analysis, the patients that did not have a GTR didn't benefit from the TMZ ... I wonder if it only works when there is small volume dz.

S

 

[medgator]

There are trials looking at temodar with Grade IIIs, trying to stratify based on chromosomal status, but IRL, most of these patients receive temodar anyway along with ~54-60 Gy of XRT after resection.

It's hard to disagree with this approach. Seriously, PCV is quite toxic, and from a QOL standpoint alone, temodar is probably the way to go. When PCV was still being used, there was only cursory understanding of which patients do better (i.e. the ones with the 1p/19q deletion), and based on what I've read, it seems that the chromosomal status had more to do with these patients doing well than PCV did.

It was pretty lame of the stupp trial to make RT alone the comparison arm (a similar gripe with the erbitux HNC trial) rather than RT with BCNU, but what can you do?

 

[G'ville Nole]

Eh... Why not?

Vitamin T is not very toxic, and there is a clear benefit for GBMs, so I can't see why not extrapolate for adjuvant treatment of AA, regardless of oligo component or MGMT/1p19q status. Even for lower grade, no data, but nothing really works so if you're for some reason treating with RT post-operatively (large tumor, crossing midline, recurrence, whatever), why not just add it? I can't see it hurting and nothing else really helps.

S

Respectfully, Simul, the reason I'd approach this with caution is that they are different diseases. We have a relatively robust understanding of the differences in molecular pathways between de novo and secondary (i.e. progression from low to high grade) GBMs. One key difference between the two is the higher proportion of p53 mutations in secondary GBMs, which are most commonly the result of specific point mutations in the p53 codons. Further, these types of mutations appear to be more prevalent in gliomas that have a silenced MGMT gene, meaning that, in theory, the use of alkylators in these types of tumors may actually drive the p53 mutation process by crippling the cell's ability to repair the point mutations.

Do we see this bearing out clinically? I agree that this is less clear, although an older analysis of RTOG studies by George Laramore showed a median survival decrement in AAs when alkylators were used (median survival 3 vs 2.3 yrs). Hopefully, the results of RTOG 9813 will give some further guidance, but on my oral boards, my answer will be "outside the confines of a clinical trial, I'd use RT alone".

In a related story, everyone out here in the community with a glioma gets temozolomide, regardless of grade.🙂

 

[deleted4401]

Makes sense.
I stand corrected, especially for orals.
But, I think in practice, I'd go by gut and give it, evidence be damned 🙂
S

 

[Gfunk6]

Respectfully, Simul, the reason I'd approach this with caution is that they are different diseases. We have a relatively robust understanding of the differences in molecular pathways between de novo and secondary (i.e. progression from low to high grade) GBMs.

I don't think the pathway has been dissected as well as, say, familial colon cancer. Gliomas in general are still highly heterogeneous though some mutations appear to be more common than others. (PMID: 17022734).

One key difference between the two is the higher proportion of p53 mutations in secondary GBMs, which are most commonly the result of specific point mutations in the p53 codons. Further, these types of mutations appear to be more prevalent in gliomas that have a silenced MGMT gene, meaning that, in theory, the use of alkylators in these types of tumors may actually drive the p53 mutation process by crippling the cell's ability to repair the point mutations.

Don't get me wrong, I'm a lab guy and I love this stuff but this data is too weak to base clinical decisions on. Without micro-array analysis of resected tumor specimens I don't think we can make accurate, blanket statements about anaplastic astrocytomas.

Do we see this bearing out clinically? I agree that this is less clear, although an older analysis of RTOG studies by George Laramore showed a median survival decrement in AAs when alkylators were used (median survival 3 vs 2.3 yrs). Hopefully, the results of RTOG 9813 will give some further guidance, but on my oral boards, my answer will be "outside the confines of a clinical trial, I'd use RT alone".

That study by Laramore analyzed IV nitrosureas (e.g. BCNU) as the chemo of choice. These are much more toxic than oral temozolomide and I don't know if it is a fair comparison. Also non-randomized data seems to indicate that oral temozlomide is equivalent to PCV in terms of PFS/OS w/ the obvious benefit being significantly less toxicity and better patient convenience.

I think your potential oral boards answer is a good, cautious one however though an equivalent argument could be made for use of temozolomide as well. From talking to senior residents, it seems that for oral boards the key is to have an evidence-based, well-thought out answers. Sometimes there is no right or wrong.

 

[G'ville Nole]

GFunk, I agree with with your counterpoints about the strength, or lack thereof, of the basic science data, and with the limitations of the older study. My intent was to advocate for a bit of caution when extrapolating the use of TMZ across a broad spectrum of gliomas, especially when they are, as you said, a heterogenous group of tumors. The mechanism I described is not an original creation, since I'm not a basic science guy and not smart enough to hypothesize such a thing (I think, but am not positive, that the hypothesis is attributed to Arnab Chakravarti).

At any rate, I agree that I wouldn't use this as a basis for clinical decision making. I would however use the paucity of prospective data concerning TMZ/XRT for Grade II or III gliomas in the primary setting as a rationale to consider XRT alone.

I'll also concede two points about concurrent TMZ: 1) It at least shows activity and has a reasonably benign toxicity profile, so I won't really object if my med oncs want to use it, and 2) If I did object, and pointed out the lack of strong data to my referring med oncs, they'd probably think I was an obnoxious snot-nosed kid straight out of residency and stop sending me patients.

I think your potential oral boards answer is a good, cautious one however though an equivalent argument could be made for use of temozolomide as well. From talking to senior residents, it seems that for oral boards the key is to have an evidence-based, well-thought out answers. Sometimes there is no right or wrong.

True dat. A good phrase to have handy is "This is a controversial area, but I would recommend x..." and then have some data to support x. As long as you recognize the controversy, no one will fail you, even if they disagree with you. Well, not unless they're an a-hole.

 

[clintpark]

I believe Dr. Minesh Mehta said that he uses XRT alone for WHO III AA and reserves TMZ for recurrence, for which there is good evidence that it's helpful. G'Ville, please verify...

In practice, all medical oncologists that I personally worked with use Stupp regimen (with up to 12 months of adjuvant TMZ).

 

[G'ville Nole]

I believe Dr. Minesh Mehta said that he uses XRT alone for WHO III AA and reserves TMZ for recurrence, for which there is good evidence that it's helpful. G'Ville, please verify...

In practice, all medical oncologists that I personally worked with use Stupp regimen (with up to 12 months of adjuvant TMZ).

You are correct. I had Minesh specifically in mind when formulating my oral board answer. You're also correct about TMZ in the recurrent setting. There's a pretty good body of literature on this, and in reality much of the response data in AA comes from these trials, in part leading to the TMZ arm of 9813.

BTW, I love these exchanges! This is one thing I miss about being away from an academic center.

 

[Palex80]

Makes sense.
I stand corrected, especially for orals.
But, I think in practice, I'd go by gut and give it, evidence be damned 🙂
S

And that's pretty fine as long as Temozolomide is well tolerated.
But there are patients out there that have serious Temozolomide related toxicity, I have even seen one die because of Temozolomide because of protracted heavy thrombocytopenia.

The "gut" feeling may be right, but we have to be careful not to burden our patients too much with non-evidence based approaches that may contribute little to their survival and simply cause side effects.

On a side note:
The reason many physicians prescribe temozolomide to "everyone with a glioma" may also be company related. The patent of temozolomide wont last forever and the pharmaceutical firm is pushing it quite actively IMHO.
And simply because it's quite easy to administer, has little toxicity, etc you may see non-oncologists prescribing it (for example neurosurgeons).

 

[TarHeelRadOnc]

There was an abstract this year about the occurence and incidence of high grade hematologic toxicity with Temodar. Definitely infrequent, but will give me pause in the routine use of concurrent TMZ for grade II gliomas in my practice. See below:

http://www.asco.org/ASCOv2/Meetings/Abstracts?&vmview=abst_detail_view&confID=65&abstractID=35054

 

[medgator]

There was an abstract this year about the occurence and incidence of high grade hematologic toxicity with Temodar. Definitely infrequent, but will give me pause in the routine use of concurrent TMZ for grade II gliomas in my practice. See below:

http://www.asco.org/ASCOv2/Meetings/Abstracts?&vmview=abst_detail_view&confID=65&abstractID=35054

I can't imagine anyone using it for Grade II's. Now Grade III's OTOH.

 

[TarHeelRadOnc]

"Vitamin T is not very toxic, and there is a clear benefit for GBMs, so I can't see why not extrapolate for adjuvant treatment of AA, regardless of oligo component or MGMT/1p19q status. Even for lower grade, no data, but nothing really works so if you're for some reason treating with RT post-operatively (large tumor, crossing midline, recurrence, whatever), why not just add it? I can't see it hurting and nothing else really helps."

I was responding to the comment above. Interestingly, however, post-RT PCV was almost significant for OS in high risk LGG (age <40 w/ STR/Bx only or age >40) in RTOG 98-02 (HR 0.72, p=0.13 log rank) and was significant for DFS (HR 0.60). And among 2yr disease-free survivors, both endopoints were significant (HR 0.52 and 0.45, respectively). \

I would use TMZ for AA as well. Not included Stupp, but included in prior trials of nitrosureas. Suprised that 98-13 hasn't been reported...

 

[Raygun77]

Respectfully, Simul, the reason I'd approach this with caution is that they are different diseases. We have a relatively robust understanding of the differences in molecular pathways between de novo and secondary (i.e. progression from low to high grade) GBMs. One key difference between the two is the higher proportion of p53 mutations in secondary GBMs, which are most commonly the result of specific point mutations in the p53 codons. Further, these types of mutations appear to be more prevalent in gliomas that have a silenced MGMT gene, meaning that, in theory, the use of alkylators in these types of tumors may actually drive the p53 mutation process by crippling the cell's ability to repair the point mutations.

Nice discussion,

I'm curious as to whether any studies do indeed stratify primary vs secondary GBMs? I'm in the middle of pushing protocols through for a prospective look at patterns of GBM progression (with FET-PET being used to predict sites of recurrence) following the stupp regiment, and it'd be interesting to stratify for primary vs secondary GBM, i think.

 

[Palex80]

Nice discussion,

I'm curious as to whether any studies do indeed stratify primary vs secondary GBMs? I'm in the middle of pushing protocols through for a prospective look at patterns of GBM progression (with FET-PET being used to predict sites of recurrence) following the stupp regiment, and it'd be interesting to stratify for primary vs secondary GBM, i think.

I think your idea makes sense.
On a side note: If you are going to treat non-resected GBMs within this protocol, then maybe it would be prudent to watch out for °II-°III gliomas that have only focally transformed into a GBM. Some authors suggest using FET (or MET) PET or MR-spectroscopy to identify which parts of the glioma have transformed. Another idea would be to match biopsy with MRI images and see which parts of the tumour were actually biopsied, or even perform several biopsies.
I remember of a patient I treated 4 years ago with a focally transformed non-resected GBM. She is still alive and showing almost no progression, probably because only a small part of the tumour transformed into a GBM and maybe (now I am speculating) the parts that transformed were still in some "early" stage of transformation and therefore not very aggressive yet. Her glioma was quite near the brain stem and we had to lower the dose to the medial parts of the tumour, which looked like still °II glioma in PET and MR-spectroscopy to <54 Gy.

 

[Raygun77]

I think your idea makes sense.
On a side note: If you are going to treat non-resected GBMs within this protocol, then maybe it would be prudent to watch out for °II-°III gliomas that have only focally transformed into a GBM. Some authors suggest using FET (or MET) PET or MR-spectroscopy to identify which parts of the glioma have transformed. Another idea would be to match biopsy with MRI images and see which parts of the tumour were actually biopsied, or even perform several biopsies.
I remember of a patient I treated 4 years ago with a focally transformed non-resected GBM. She is still alive and showing almost no progression, probably because only a small part of the tumour transformed into a GBM and maybe (now I am speculating) the parts that transformed were still in some "early" stage of transformation and therefore not very aggressive yet. Her glioma was quite near the brain stem and we had to lower the dose to the medial parts of the tumour, which looked like still °II glioma in PET and MR-spectroscopy to <54 Gy.

Thanks Palex!

The heterogenous nature of glioma makes things so much more complicated, eh! For my retrospective analysis I'll definitely keep that in mind- I'm sure I'll be able to dig up path reports specifying if the GBM was focal or widespread relative to the whole tumour. And good to hear a success story! Just proves that for all our data- nothing in oncology is EVER certain.

In terms of performing biopsies to serve as the gold standard, ethics is very difficult (since it's a very invasive procedure that may not be indicated!) and you'd need the neurosurgeon on board...tough as well!

Our current prospective study design has patients treated with the standard C+ MRI plan, but with fused FET-PET/MRI GTVs made up as well, separately.

Patient outcome will be correlated to whether or not FET/MRI GTV was fully enclosed by the clinical treatment volume (as per C+ MRI with standard 2-3cm margins) and sites of tumour progression (imaging, biopsy with clinical criterion) will be correlated to FET GTV regions that were or were NOT in the CTV- we're thinking more failure will be non-local if this is the case!

Pretty technical stuff, but hopefully should shed some light on HGG progression patterns, as well as the potential use of FET-PET in RTP!

 

[Palex80]

Patient outcome will be correlated to whether or not FET/MRI GTV was fully enclosed by the clinical treatment volume (as per C+ MRI with standard 2-3cm margins) and sites of tumour progression (imaging, biopsy with clinical criterion) will be correlated to FET GTV regions that were or were NOT in the CTV- we're thinking more failure will be non-local if this is the case!
Pretty technical stuff, but hopefully should shed some light on HGG progression patterns, as well as the potential use of FET-PET in RTP!

Perhaps we will have to go back to WBRT one day, especially if more trials with local dose escalation provide enhanced local control.

 

[Raygun77]

Perhaps we will have to go back to WBRT one day, especially if more trials with local dose escalation provide enhanced local control.

thats the confusing thing though! One trial with a proton boost to 90Gy equivalent showed excellent local control, however others don't, for example this study:http://www.ncbi.nlm.nih.gov/entrez/...ve&db=pubmed&dopt=Abstract&list_uids=11896114

still local failure, with 90Gy! I'm curious as to the occurrence of radionecrosis at 90Gy as well...

 

[Palex80]

thats the confusing thing though! One trial with a proton boost to 90Gy equivalent showed excellent local control, however others don't, for example this study:http://www.ncbi.nlm.nih.gov/entrez/...ve&db=pubmed&dopt=Abstract&list_uids=11896114

still local failure, with 90Gy! I'm curious as to the occurrence of radionecrosis at 90Gy as well...

1. Some speculate that GBM cells migrate through the brain away from the primary tumour site and return back to it after therapy is over, thus creating a recurrence. That's the reason why theoretically WBRT may be of advantage in the age of Temozolomide and local dose escalation.

2. Radionecrosis after 90 Gy may very well be an issue. You may come up with some data on radionecrosis after high dose radiation therapy by looking into patients treated with a brachytherapy boost (either LDR or HDR).

 

[Charles_Carmichael]

I apologize for bumping an old thread but, since there's already a good discussion here, I thought it would be better than starting a new thread.

Regarding GBM, I recently came across two commentaries (PMID: 18833584 and PMID: 20151424) comparing the Stupp trial with the NOA-01 trial (PMID: 12947063) that was published in JCO back in 2003. After reading through the paper, which boasts pretty impressive results ("Median survival and 2-year survival rates were 17.3 months and 25% for ACNU plus VM26, and 15.7 months and 29% for ACNU plus Ara-C in glioblastoma"), I'm curious as to why this didn't receive more attention than the Stupp trial.

I understand that you can't really compare across trials but it appears to me that the NOA-01 trial produced superior results compared to the Stupp protocol. The first commentary I linked from 2008 (PMID: 18833584) has a nifty chart comparing patient characteristics, OS, etc, from both the trials if you're interested. There doesn't appear to be a significant difference in the patient populations between each study. The NOA-01 trial did have fewer patients enrolled in each arm and didn't have a RT-only arm, but I can't seem to see any major flaws (compared to the Stupp study) that would question the results.

So, what exactly am I missing that makes the Stupp protocol superior to the ACNU + teniposide arm of NOA-01? There must be a good reason as to why the Stupp protocol is the current gold standard for GBM. Are there actually big differences in patient populations that I'm missing (due to not being a clinician)? The toxicity profile doesn't seem significantly different and recent preclinical data (I haven't really looked into clinical data yet) suggest that ACNU has the most favorable toxicity profile compared to the other nitrosureas. I'm also a little surprised that the ACNU + teniposide regimen hasn't been looked into further or compared with the Stupp protocol considering the favorable results. Thoughts?

 

[Gfunk6]

So, what exactly am I missing that makes the Stupp protocol superior to the ACNU + teniposide arm of NOA-01? There must be a good reason as to why the Stupp protocol is the current gold standard for GBM. Are there actually big differences in patient populations that I'm missing (due to not being a clinician)? The toxicity profile doesn't seem significantly different and recent preclinical data (I haven't really looked into clinical data yet) suggest that ACNU has the most favorable toxicity profile compared to the other nitrosureas. I'm also a little surprised that the ACNU + teniposide regimen hasn't been looked into further or compared with the Stupp protocol considering the favorable results. Thoughts?

The review you quoted summarizes the reasons quite well:

1. Nitrosureas (mainly BCNU) for treatment of high-grade gliomas are treatment paradigm many clinicans do not wish to revisit
2. There was no statistically significantly difference in OS between the arms of the NOA-01 trial.
3. MGMT hypermethylation has been far better validated for temozolomide.
4. Comparing absolute numbers between two trials is a questionable practice.

If a trial is run between Stupp and NOA-01 that would be the ultimate proof.

Also, to answer Palex's original question about what do do for pts with Grade III gliomas . . . I think the best solution would be to enroll them on RTOG 0834.

 

[Island Rad]

For Grade III patients with patients anaplastic oligodendroglioma or oligoastrocytoma treated with adjuvant chemo (especially when they have MGMT+ or LOH1p/19q) what are the senior members seeing as the practice patterns when there is recurrence documented on imaging (MRI spectroscopy)? Biopsy/re-resection for path, additional chemo, RT, combination CRT?

​

 

[Charles_Carmichael]

Thanks for the response Gfunk! 🙂

1. Nitrosureas (mainly BCNU) for treatment of high-grade gliomas are treatment paradigm many clinicans do not wish to revisit

Yea, I can understand that. A lot of the older literature was with BCNU/CCNU rather than ACNU though. I could be wrong, but I do remember reading some literature suggesting that the efficacy and toxicity profile for ACNU is better than that of the other nitrosureas.

2. There was no statistically significantly difference in OS between the arms of the NOA-01 trial.

Agreed. The main difference seems to be the toxicity profile of each arm.

3. MGMT hypermethylation has been far better validated for temozolomide.

That's true; there's a fair bit of literature suggesting that MGMT methylation status is predictive of response to alkylating agents. Some recent evidence seems to suggest that MGMT hypermethylation may be a favorable prognositic factor regardless of treatment though. I guess we just have to wait and see if that pans out or not.

4. Comparing absolute numbers between two trials is a questionable practice.

Definitely agree. I was a little hesitant to bring up this topic because of that.

If a trial is run between Stupp and NOA-01 that would be the ultimate proof.

That would be pretty awesome but I can't see it happening. Anyways, thanks again for the response. Appreciate it.

 

[Palex80]

Temozolomide definetely is a very "convenient" drug. It's an oral agent, patients take at home. Other types of chemotherapy are administered i.v., which is a major drawback for these patients, who have a limited prognosis.


Other than this, I fully agree with you Kaushik. We don't have hard evidence demonstrating the supremacy of Temozolomide over other treatments and we cannot definetely say, that the trials carried out before the Stupp trial were powered enough to demonstrate a relevant overall survival benefit.


Ultimately what we need is a randomized trial looking into different chemotherapy combinations. However this is not the way current research is going into.