Wednesday, 10 May 2017

Oligoastrocytoma grade 3, progression, next steps?

Hello all and thank you for this great community!

My sister (30yrs) was diagnosed with anaplastic oligoastrocytoma GIII a year ago. This was a reoccurence, since 5 years ago it was oligoastrocytoma grade 2. Back then she received surgery + 8 cycles of TMZ.
This latest one was partially removed in surgery (it's in temporal lobe). She received RT + TMZ (60Gy, 30days) and after that she has completed 10 TMZ cycles. Latest MRI showed that there might be tumour progression, but it's not definite according to doctors. Next scan will be in two months. Until then she will continue on TMZ for two cycles more.

We're a bit concerned since part of the tumour is still there even after RT and chemo and it might be progressing.
PAD report says: IDH1 negative, mitosis 33/10 HPF, MIB activity 25%,
Previous tumour showed 1p19q deletions.
We have asked for a more specific analysis of the tumour and will receive results in couple of weeks (MGMT, EGFR, etc.).

Until then, our NO recommends to continue with TMZ and see what MRI shows in July.

We have plans to try Keytruda or other aPD1, what's your view? Or is there some other sytostate (CCNU, PCV) we should try before aPD1 ?

Her clinical condition is very good so we try to balance between good quality of life and try to treat this when there's better changes for that.

Supplements she currently takes:
-Longvida Curcum
-Probiots (L. Casei et al.)

br and thanks,


  1. Hi Juha, maybe you could consider the ketogenic diet as well.

    1. Thanks Anna! We'll look more into ketogenic diet. It'd seem like a good option at this point.

  2. We live in Europe, do you know if there'd be any suitable clinical trials ? The most promising ones I've found so far has been in US.

  3. Have you thought of any of the German clinics. We use IOZK in Cologne under Prof Stefaan van Gool since January 2016

    1. Hello Alice!
      Yes, we have thought about it and it definitely seems very interesting. Of course it's very expensive since we have no insurance.
      But there are couple of concerns still:
      -the surgery was done a year ago so the biopsy sample is not fresh and tumour might have developed after a year
      -IOZK said that they could make a vaccine out of a blood sample, but I'm a bit sceptic would this actually work, since tumour markers in blood have very low concentration to my understanding.

      Alice, have you been on DCVax + (Keytruda/Opdivo) ? Any side effects ? Has there been good respond to treatment? Did you do any analysis of the tumour beforehand regarding immunological markers ?

    2. Hi it was for my Dad. he had fresh tumour from the >90% resecction and had two vaccinations last July and August. He has been going monthly since January 2016 every 8 weeks, for the NDV Vit C selenium and hyperthermia treatment. he is very well, no side effects, back at work, and and is also taking the cocktail, supplements and thc/cbd.

    3. I would certainly discuss with prof Van Gool. He is very patient centered and will discuss by skype.

    4. Thank you Alice,

      Good to hear that your father's doing fine.

      Sounds promising,
      I'll contact IOZK-clinic next week ans discuss in more detail with them!

    5. Do you know other patients using vaccine treatment effect

  4. Juha,
    I'm puzzled by the 1p/19q codeleted status of the original tumor, but the IDH1 (mutation) negative status of the new tumor. Was the original tumor tested for IDH1 status? Even if testing negative for IDH1 R132H originally, there could be alternative IDH1 mutations (IDH1 R132__), or IDH2 mutation which is more common in oligodendroglioma than astrocytoma.

    If this is an IDH-mutant, 1p/19q codeleted tumor, it would be classed as an oligodendroglioma under the updated classification system. If so, I'd be concerned about hypermutation due to the multiple previous rounds of TMZ. And if hypermutated, a PD-1 or PD-L1 antibody would be a good idea. This is a lot of "ifs". Would be good to get some more clarification on the IDH1/2 status of the original tumor.

    1. Hi Stephen,
      I found the PAD-report from the orignal tumour: It was also IDH1-negative. 1p36 and 19q13 deletions found in appr. 38% of the cells. MIB proliferation was 2-3% back then.

      We should receive detailed analysis of the latest tumour in couple of weeks (OncoDNA-analysis).

      Is it so that oligoastrocytoma with IDH1-negative status is close to GBM ? This is something we're very afraid of.

    2. Excellent, I'm glad to hear you're getting the OncoDNA analysis. That should conclusively tell you whether or not there's an alternate IDH1 mutation or an IDH2 mutation. Until IDH1 and IDH2 are sequenced, we won't know for sure.

      I know of several patients personally, whose tests originally showed no IDH1 mutation, but after sequencing they were found to have an alternate IDH1 mutation other than the more common IDH1 R132H mutation. We can't come to any conclusions until the OncoDNA testing comes in.

    3. Furthermore, the OncoDNA panel of tested genes that I'm looking at (OncoDEEP clinical +) also includes mismatch repair genes such as MSH2 and MSH6. You may get a good idea of whether the tumor is hypermutated or not when the results come in.

    4. Thank you Stephen!
      Yes, there's at least MSH2 gene on the list, I'm not sure on MSH6 (Onco Strat&Go)
      I'll get back on this when we get the results.

    5. Hello Stephen and others! Now I have the result from OncoDNA-analysis:
      Analysis main points:
      -TP53 mutated (44%)
      -IDH1 mutated (45%)
      -MGMT methylated
      -EGFRvIII not mutated
      -CD8 negative expression
      -MSH2 not mutated
      -TOPO1 & TOP2A positive expression
      -phosphoRb positive expression
      -cMET negative expression
      -pERK1/2 low expression
      -PD-L1 low expression
      -p4EBP1 high expression
      -No CopyNumberVariations found (CNVs)

      As a summary, it seems that tumour has very small number of mutations (according to our NO it's "cold-tumour"). This together with negative PD-L1 means that aPD-1 - drugs might not the best choice. Although it may be that RT + TMZ have changed that. Sample was taken before RT+TMZ (a year ago).

      Irinotecan/etoposide might work (topoisomerase inhibitors). Also palbociclib (CDK4/6) inhibitor) was suggested.

      So far, we have completed 11 rounds of TMZ, and wait next MRI to be taken in couple of weeks, but according to last MRI it seems that TMZ is not working, there might even be progression. Despite the fact that MGMT status gives positive signal for TMZ.

      We would really appreciate to have your opinions what to try next ? We have mixed feelings regarding e.g. irinotecan based on literature we've found. Sometimes it has worked, sometimes not, and normally it's combined with Avastin.

    6. Thanks for sharing the results Juha. The list you shared included only 4 sequenced genes, and 8 protein markers, plus one gene (MGMT) tested for methylation status. Was this just a short summary of the results or were these the only genes and protein markers mentioned in the report? If only a small number of genes are sequenced then only a small number of mutations can possibly be found and this isn't conclusive evidence for the presence or absence of hypermutation.

      It's very interesting, and fortunate, that IDH1 mutation was found, where no IDH1 mutation was found in previous testing. Was the IDH1 mutation an R132x mutation (with the "x" being a letter other than H)?

      The presence of TP53 mutation, but no copy number alterations (such as 1p/19q codeletion) point more towards an astrocytoma than an oligodendroglioma.

      The "cold tumor" comment refers presumably to the absence of CD8 expression (a marker of cytotoxic T cells), and low expression of PD-L1. It may therefore be true that a PD-1 antibody as a single agent might not be the best option.

      As you mention these results are a snapshot of the tumor a year ago before the radiation and 10 months of TMZ. Much could have changed since then: TMZ could have caused more mutations, the protein markers could be different now compared to a year ago, the combined treatment (especially radiation) could have caused more infiltration of immune cells. It's definitely not an ideal situation (using testing on a one year old pretreatment sample to try to predict which therapies to use today).

      Palbociclib was surely suggested because of the presence of phosphoRB, the inactivated form of the tumor suppressor RB1. Palbociclib is a CDK4/6 inhibitor - CDK4 and CDK6 are involved in phosphorylation/deactivation of RB1. The problem with CDK inhibitors is 1) cost (who would pay for this?) and 2) limited blood-brain barrier crossing. As discussed previously on this blog (see the "abemaciclib" label) abemaciclib seems to have improved penetration across the blood-brain barrier compared to palbociclib, though not yet FDA approved.

      If we're talking expensive oncology drugs, PARP inhibitors would also be worthy of consideration, especially for contrast-enhancing tumor with disrupted blood-brain barrier. See the recent discussions on the blog concerning PARP inhibitors. Recent studies have shown particular sensitivity to PARP inhibitors for IDH1-mutant tumors. Combining PARP inhibition with chemo could be effective.

      Assuming the tumor is a IDH1 mutant, TP53 mutant astrocytoma, there is also a high likelihood of ATRX mutation and loss of function, which could render the tumor more sensitive to irinotecan than the average GBM.

      Treatment in this case should proceed according to whether the tumor is hypermutated or not. The "cold tumor" comment refers to the tumor one year ago at the time of the resection, and doesn't necessarily reflect the current situation. If hypermutated, PD-1 or PD-L1 therapy would be a worthwhile option.

      If not hypermutated, my suggestion would be some form of chemo [CCNU/lomustine or PC minus the Vincristine, or perhaps irinotecan though there's much less evidence of benefit for lower grade gliomas compared to lomustine or PCV]
      + PARP inhibitor [one of the approved ones, not veliparib]
      + disulfiram/copper
      + fish oil high in DHA and EPA.

      high dose tamoxifen ala Ben Williams would also be worthy of consideration.

    7. Thank you Stephen for very detailed answer!
      Over 200 genes were analysed, but only two of them were found mutated (IDH1 and TP53).
      -"Was the IDH1 mutation an R132x mutation (with the "x" being a letter other than H)?" Yes, it was R132G mutation. The previous analysis was made by staining method, so probably that's the reason for different result.
      It seems that ATRX was not included in analysis, unfortunately.
      I didn't realize that "no CNVs" means that 1p/19q is not codeleted. First tumour occurence (2004) showed 1p/19q codeletion. Is it possible that tumour has evolved so that 1p/19q codeletion is lost ?
      Anyway, thanks for your answer, I'll need to look into all the details!

    8. What is the difference between TMZ and e.g. CCNU, they're both alkylating agents? Could CCNU work even if next MRI would show that there's progression with TMZ? And is there a risk for hypermutation similar to that with TMZ?

    9. Also, is there a major difference between other topoisomerase inhibitors doxorubicin/epirubicin/etoposide compared to irinotecan?

    10. Hey Juha, for CCNU vs TMZ check out this thread

      or google some other ones. In short TMZ and CCNU damage to the dna are repaired in different ways and so carry somewhat different mutation risks. It also means that it is possible for CCNU to work once TMZ has stopped working, but certainly no guarantee. Cancers can also end up fully skipping the checkpoint phase of DNA synthesis and then it wouldn't matter which chemo was used, the cell would go ahead and divide anyway, albeit with lots of damage to its dna.

      The major thing to consider with CCNU is that it is much harder on blood counts and thus hard on the immune system. So it carries additional risk.

    11. Another link with some info on CCNU

    12. You stated earlier that "1p36 and 19q13 deletions found in appr. 38% of the cells."
      This does not prove that these were whole-arm deletions (deletion of one copy of the entire 1p and 19q arms of chromosome 1 and 19). These could have been partial deletions only. True codeletion that defines molecular oligodendroglioma is loss of the entire 1p and 19q chromosome arms.

      "Copy number variants" refers to extra copies (more than two) or deletions (of one or both copies) of parts of chromosomes or even whole chromosome arms, as in the case of true 1p/19q codeletion. The presence of a TP53 mutation combined with no copy number variations strongly suggests to me an astrocytoma rather than oligodendroglioma. ATRX status would be even more definitive though.

      Anthracyclines such as doxorubicin are topoisomerase II inhibitors, as is etoposide, while irinotecan is a topoisomerase I inhibitor. The main challenge with chemotherapies such as these is getting them across the blood-brain barrier into the tumor. This is why alkylating agents such as TMZ and nitrosoureas (CCNU, BCNU etc) have become the primary chemo for glioma, as they penetrate into the central nervous system better than most other chemos.

      Yes TMZ and CCNU have different mechanisms of action, so its possible CCNU would work even if TMZ doesn't. There are couple of reasons to believe that CCNU would not be as likely to cause hypermutation: one reason is because its mechanism of action is different with regard to the necessity of active mismatch repair, and the other is that out of 8 oligodendrogliomas treated with nimustine (ACNU, an alternative to CCNU used in Japan), none of them developed hypermutated recurrences.
      If ACNU was as prone to cause hypermutation as TMZ, we would expect at least several of these tumors to have developed hypermutation. I'm assuming here that CCNU would be similar to ACNU in terms of this risk.

    13. Thank you both bs1966 and Stephen for detailed answers!

    14. Stephen,
      Do you know if radiation therapy is able to change permeability of BBB ? Our NO said that it might be the case and therefore topoisomerase inhibitors could have a change (and also other drugs).

    15. I found this article (PMID: 23898097) saying that etoposide improves survival in HGG whereas irinotecan might result in inferior outcome.

  5. Hi Juha,

    I can't answer your questions but I wanted to share that I reached out to Lanette from survivor stories. She is a 19 year survivor of a grade 3 Oligo. She responded quickly and was really kind. Maybe your sister could use the hope right now too. Hope is a very important part of the formula. She had a recurrence 14 years ago. It was inoperable and treated with radiation and TMZ.


    1. Hi Maria,
      Thank you very much. I'll pass this information to my sister. We try to keep our minds on positive and survivor stories are very good source for that :)