Hello all,
My sister's tumor (AA3) has progressed and a surgery is scheduled.
Would you do tumor analysis and if so, which one:
-Caris Molecular Intelligence
-Foundation One
-OncoDna?
We already know her tumor is IDH1 and TP53 mutated and MGMT methylated. No other mutations we're found last time. Also CD8+, PD1 and PD-L1 were negative. I don't know if there's much to gain this time, because it seems that the number of suitable chemos is anyway quite limited and PD1-inhibitors have shown low efficacy on IDH1mutated tumors?
And do you see benefit to store tumor sample for later use (e.g. Vaccine)?
Br,
Juha
Hi Juha - Not sure where to advise you on genetic screening but has your surgeon discussed Gliadel wafers at all? I'm not sure if used for AA3 or just GBM, but I have a friend who has them (she has GBM) and she has been stable for 2 years with nothing else. The Gliadel rep told me not every surgeon uses them, but worth the ask. Best of luck with the surgery.
ReplyDeleteAnnie
Hello Annie,
DeleteNo, our surgeon hasn't mentioned Gliadel wafers, but I think they're not available in Finland (Europe). I'll double check that, but thanks anyway!
I don't know about the others, but Foundation One will provide you with a list of drugs they calculate are likely to have an effect. No one knows how good the list is though.
ReplyDeleteStoring the tumor material is definitely wise. Ask what method they will use for storing -- I believe freezing is the way to go. Where will the surgery be done? Any chance they are using the new "tumor paint"? http://bit.ly/2xHpK9v Wishing your sister a safe and successful surgery.
ReplyDeleteThanks David!
DeleteYes, we would go with cryo-preservation, probably with StoreMyTumor-company.
Unfortunately we don't have the option for "tumour paint". The surgery is done in Finland.
Hi Juha,
ReplyDeleteStandard procedure is to store tumor material, but in paraffin, vs flash freezing as David is noting. Genetics can be done with paraffin, most viral/immuno stuff needs flash frozen. Not all surgeons will even do flash frozen and I think it is often challenging to get things sorted to freeze and store AND to have full rights to the tissue (ie to be able to take it all and use it with another hospital/university/etc). It will also only be useful if you can enroll in a clinical trial or pay out of pocket to one of couple of clinics that are doing immuno (mainly in germany).
In terms of who to pick or value of it, the major complicating factor is that IDH1 mut tumors often don't have that many mutations, but DO have huge amounts of methylation damage from 2-hg. So in a lot of ways it may be more important to look at tests that include epigenomic tests, not specifically DNA testing. For instance MGMT testing is sort of checking how active or inactive is MGMT. It could be totally mutated or more likely the site is methylated and has reduced expression.
So maybe you can actually get the testing companies to explain what their test suite will actually provide beyond straight genetics. Note that drug suggestions may not be that useful because I doubt they parse out whether or not the drug penetrates the BBB. But maybe I'm wrong about that.
Also per David's suggestion on 'tumor paint', 5-ALA has just gotten FDA approval, and is a similar kind of tool that highlights tumor cells and improves resection success. Its probably in the category of a must have technique.
One other thought is the Toca 5 trial would be an option. The results for IDH1 mut were very good, the major downside is that it has a 'control' arm though it is unclear if it is a fully blinded study. Its enrolling at many institutions in the US and Canada. But not sure where you are located.
Bryan
Thanks for your always informative and thoughtful posts Bryan. A few things that I would add:
DeleteIDH mutant gliomas virtually always display the G-CIMP (glioma CpG Island Methylator Phenotype), which as you correctly note is the result of 2-HG inhibiting enzymes involved in cytosine demethylation (such as TET2).
I think the methylation/epigenomic testing you were referring to was of this type:
https://cancergenome.nih.gov/abouttcga/aboutdata/platformdesign/illuminamethylation450
The Costello lab at UCSF has done some important work showing these IDH1-mutant, G-CIMP gliomas can lose methylation at genes involved in the cell cycle, leading to their increased activation.
https://www.ncbi.nlm.nih.gov/pubmed/26373278
However I think the testing for hypermutation ("tumor mutational burden" - FoundationOne, or "total mutational load" - Caris) is the most critical test for recurrent IDH1 mutant glioma, as it can dictate the type of chemotherapy that should be attempted, or the likelihood of PD-1/PD-L1 antibodies being useful.
Totally agree about the drug suggestions not being that useful. I usually pay little attention to those recommendations when reading the reports.
There was a recent study showing IDH1 mutant tumors were much less likely to fluoresce after 5-ALA, but the study was a mix of all different grades and doesn't address the question whether this outcome was due to the IDH1 mut tumors being more often lower grade and less often GBM.
https://www.ncbi.nlm.nih.gov/pubmed/28740449
Tumor Paint refers specifically to BLZ-100, which is only in the early clinical trial stage, and none of the trials are currently recruiting.
https://clinicaltrials.gov/ct2/show/NCT02234297
https://clinicaltrials.gov/show/NCT02462629
As you noted some of the 5-ALA studies have a bit of a flaw in them in regard to grade. Basically some newer studies and clinical trials are indicating that when you combine 5-ALA with high magnification surgical microscopes that it is possible to see the fluorescing regions, even in low grade oligo (allmost all IDH1 mut). Basically showing that 5-ALA works, but lower cancer cellularity cases make it harder to see. Unfortunately for the time being to take full advantage of 5-ALA if the tumor doesn't have a lot of contrast enhancment on MRI, it means it takes even more work sorting out where to get surgery to tie in with a trial of this technique.
DeleteCheck out :https://www.ncbi.nlm.nih.gov/pubmed/21761971
Another comment on the testing is that some big institutions also do their own genetic panels. UCSF has the 'UCSF-500', which had a version 2 update in 2015 and is almost 500 different genes. They don't directly compute a 'mutational burden' but list allele frequencies for each of the mutations found which allows a basic estimation. But since even 500 genes is a paltry number out of the whole genome, an actual TML calculation is certainly a different level.
Overall though, Stephen your comment about being in Europe certainly makes a lot of my comments less useful.
Even without "total mutational load" quantifications, I could often infer whether a tumor was hypermutated or not with the Foundation reports of 300 or so genes tested. If on a FoundationOne report there was a large number of mutations discovered (usually around 40-100, including the variants of unknown significance in the appendix), including a mutation in a mismatch repair gene (usually MSH6), it was pretty obvious the tumor was hypermutated. Of course the newer TML/TMB testing is a much more direct measurement that leaves the guesswork out.
DeleteThank you Brian and Stephen for very informative replies!
DeleteThe standard procedure would be to have a tumor sample in paraffin. We're now planning to have it cryo-preserved to keep all the options open after the surgery (clinical trials, vaccines, ...). It still need's to be confirmed with the hospital that the procedure is ok with them, but I don't see any problems with it as long as they can have their mandatory sample for a pathologist.
We'll discuss with analysis companies about their methods and request more information about their test suite.
Unfortunately ALA-5 or other "tumor paints" are not available at our hospital, at least not yet, hopefully in the future.
Brian: Do you know if Toca5 - trial is still recruiting? I've understood that it's not taking anymore participants.
Hi Juha,
DeleteI'm really sorry to say that I hadn't rechecked the status listed for the Toca5 trial and see now that it has been switch to 'active, not enrolling' as of a few months ago. It is probably still worth trying to contact the lead investigator because maybe they would still add in another patient.
In terms of cryo-storage, if the hospital doesn't do it, getting it done may not be that simple because of required sample prep required and limited time available before it needs to be frozen.
I don't know anything about getting into US trials if you are from Europe, but in the US depending on the trial sometimes the costs are covered by the trial and other times it is the patient's insurance that pays (ie for cases where surgery is part of the trial).
So maybe others could comment but maybe entering a US trial is an option, but maybe surgical trials are more of an issue regarding insurance. For instance the 5-ALA trial at UCSF would set you up with some real top notch surgeons with the latest equipment (https://clinicaltrials.gov/ct2/show/NCT01116661?term=5-ala+glioma&recrs=acdf&rank=4)
Hi Brian!
DeleteNo problems. I also think it's worth to try and contact Toca5 investigator, you never know :) Let's see what they respond.
Yes, we'll need to check the cryo procedure with our hospital, if it can be done and that it's done properly.
And thanks for the trial suggestions, we'll explore those.
As to which company to send away the tissue for analysis, it should include a measurement of tumor mutational burden, which would determine if the recurrence is hypermutated or not. Durvalumab (PD-L1 antibody) actually seems to work very well in recurrent IDH1-mutant GBM, and my theory is that these were actually hypermutated cases.
ReplyDeleteSee my writeup on the phase 2 durvalumab results here:
http://astrocytomaoptions.com/sno-2016-clinical-highlights-scottsdale-arizona/
FoundationOne reports include "tumor mutational burden" quantification as of 2016
http://investors.foundationmedicine.com/releasedetail.cfm?releaseid=985071
and Caris reports include this now as well
http://www.carislifesciences.com/cmi-overview/total-mutational-load-tml/
I'm not sure if OncoDNA does anything similar to this.
FoundationOne only works with USA-based insurance providers so you'd be paying entirely out of pocket if you decided to go with them. Caris is more international and has in international office so my guess is more likely that European medical insurance would cover at least part of the cost.
Thank you Stephen, we'll look into these more detail!
DeleteUnfortunately, we don't have any insurance for extra tests, so it will be out of our own pockets anyway. On the other hand, we can choose the best one for my sister's case.
Dear All,
DeletePlease, note that OncoDNA does Tumor Mutational Burden now.
You can contact them via the chat on the web site www.oncodna.com or infos@oncodna.com.
Their solutions are cheaper than Foundation or Caris :-)
If DC vaccination comes into question, then one need about 1,5 cm³ viable tumor tissue STERILE, DRY, FROZEN at -80°C.
ReplyDeleteHello SVG,
DeleteDo you know if StoreMyTumor - company fulfill these requirements? They claim to use "cryo-preserve" (http://www.storemytumor.com/services).
Do you know if there's any publications or data on how effective DC vaccines are for AA3 IDH1-mutated tumors?
After sending some emails, I found that OncoDNA offers the Tumor Mutational Burden testing as part of the OncoDEEP package, and not as a standalone test.
ReplyDeletehttp://www.oncodna.com/en/solution/oncodeep/
With Caris, it's possible to order only the Next Generation Sequencing panel (including Tumor Mutational Load), as opposed to the entire Molecular Intelligence package. The Next Generation Sequencing testing alone is $3000 USD less than the cost of the entire Molecular Intelligence package which offers many other tests, and also includes the Next Generation Sequencing.
https://www.carismolecularintelligence.com/total-mutational-load-tml/
The problem with OncoDEEP is that it only tests for 75 genomic alterations/mutations, whereas Caris tests for 500+. That is great that you can just order the Next Generation Sequencing by itself, I thought that it had to be ordered as part of the Molecular Intelligence package - which is quite prohibitive in terms of price.
ReplyDeleteFor Caris Molecular Intelligence, I just found the online document with prices for both the full Molecular Intelligence profile and for the Next Generation Sequencing only.
ReplyDeletehttps://www.carismolecularintelligence.com/wp-content/uploads/2016/12/TN0203-v3-CMI-Patient-Billing-Practices_Page_1.jpg
Down in the uninsured or self-pay section: $6500 for the full MI Profile, and $3500 for Next Generation Sequencing only.
A new version of OncoDEEP has been released. They added:
ReplyDelete1. Genes mainly dedicated to immuno and targeted therapies
2. MSI (Microsatellite Instability) testing for Immunotherapy and resistance to 5-FU based chemotherapies
3. Sequences required to perform TMB (Tumor Mutational Burden) analysis for Immunotherapy
4. Highly polymorphic SNPs to detect hemizygosity or LOH (Loss of Heterozigosity) for targeted therapies or prognosis.