Recurrence patterns after maximal surgical resection and postoperative radiotherapy in anaplastic gliomas according to the new 2016 WHO classification

Authors: Jung Ho Im, Je Beom Hong, Se Hoon Kim, Junjeong Choi, Jong Hee Chang, Jaeho Cho, Chang-Ok Suh


Scientific Reports, 2018, Volume 8, Issue 1

Why is this paper so important?

However, recurrence patterns according to the histological subtypes have not been well studied. Most studies on recurrence patterns were mainly conducted before 1995

This is the reason why this paper is important because it focuses on the type III tumours (which hasn’t been explored in contemporary literature) and the first article to address this.

I am more interested to see how the molecular landscape varies across the different histologies. As you can see from the ensuing table, majority of them were MGMT methylated with IDH1 mutation and some gratifyingly 1p19q co-deleted (for the Anaplastic Oligodendroglioma) (which are known to have a good prognosis, anyway).

The recurrence patterns are more instructive. The following is a representational image of how they took the planning volumes (but more importantly, the way recurrent tumours were defined- they were “outside” the planning volumes.

Let’s talk about the vital survival curves; there’s not much difference between the PFS and OS (which anyway is a known factor for the Grade III tumours). There is nothing earth-shattering about this, in the first instance.

But look carefully at the following table which provides the low down of the molecular abnormalities in the tumours which were identified in the cohort.

Survival outcomes as per the molecular findings; of course, this confirms our biases 🙂 As always, the gross total resection offers the best results along with the mutated IDH and co-del tumours. Again, there’s nothing much to bleat about here because these are known “good prognosis factors”.

Median PFS and OS were 130 (range, 56–204 months) and 158 months, respectively. 5- and 10-year PFS rates were 60.4% and 56.3%, respectively, and 5- and 10-year OS rates were 65.4% and 56.6%, respectively

The authors had used the antibody to test for the IDH presence; although reasonably robust but I ALWAYS prefer to get the sequencing done to identify the mutation if any.

Antibody used was anti-human IDH1 R132H mouse monoclonal antibody (Clone H09L, 1:80 dilution; Dianova, Hamburg, Germany). When the cytoplasmic expression of IDH1 R132H was identifed in glioma cells, we considered the case as “mutant”/“positive.”

They had a pretty robust way to identify the 1p/19q status as well (as per the 2016 guidelines for identification of FISH criteria).

The table is demonstrating break up of the mutations in the cohort (which I feel is the most important table in the entire paper).

THIS IS THE MOST IMPORTANT FINDING.

 

 

 

 

 

 

 

 

 

 

 

Patterns of recurrence according to the 2016 WHO classification and extent of resection. Column 2 represents the number of patients in each subgroup with recurrence and their percentages are indicated in parentheses. GTV, Gross Tumor Volume; CTV, Clinical Target Volume; CSF, Cerebrospinal Fluid; WHO, World Health Organization; AO, Anaplastic Oligodendroglioma; IDH, isocitrate dehydrogenase gene; GTR, Gross Total Resection; STR, Subtotal Resection; PR, Partial Resection; Bx, Biopsy; AA, Anaplastic Astrocytoma.

If you look carefully, the ones that had the maximum out of field recurrences were IDH wild type. Is it possible that they had some pro-migratory characteristics right from the start? Are we missing something?

Recurrence rates for GTR or STR were similar but much lower than those for those who underwent partial resections or biopsy. In patients with GTR, most recurrences (12/13) occurred in the patients with AA, IDH-wildtype. 

The authors had followed the standard protocol.

We prescribed 46–50 Gy to CTV. Therefore, a higher radiation dose to CTV can be considered for AA, IDH-wildtype

I would concur with the above statement.

The Achilles heel of the paper? They had chosen the patients selectively for getting chemotherapy. Not everyone got the adjuvant TMZ.  I prefer to give TMZ to all patients irrespective of the mutational status; however, the bigger question is whether it ought to be for 6 months, 12 months or even up to 24 months. I think, in part, it depends on the performance scale of the patients as well as the MRI perfusion scans. If the perfusion value is less, I’d prefer to keep at it. But these are only anecdotal observations; not backed by the trial outcomes. Which gives me an idea for a study as well.

The role of TMZ has been assessed in a large international trial, CATNON. Interim analysis showed that adjuvant TMZ CTx was associated with significant survival benefits in patients with newly diagnosed 1p/19q non-co-deleted anaplastic glioma.

The authors have proposed increasing the size of the planning target volume. They do mention the newer 2HG MR spectroscopy which detects the abnormality over a wider area.

Metabolic imaging with MR spectroscopy to image the 2-hydroxyglutarate signal in the brain to detect oncogenic IDH1 mutations has been recently developed. It showed that the 2-hydroxyglutarate volume was larger than the FLAIR volume in approximately half of the IDH-mutant glioma patients

But the big question: How much normal brain can you really irradiate?

These are the broader issues that need to be debated.

Is research wasted?

How far is this true?

I would dispute the numbers but the questions raised here are extremely pertinent.

  • As always, we need to ask the right question. Design appropriate methodology.
  • Put it out in public domain.
  • And make the research accessible.

These are desirable goals but not the end points that we see in practise. Things have to change for meaningful outcomes.

Research in Radiation Oncology: Unanswered questions

 

Paper identified:

Current and Future Initiatives for Radiation Oncology at the National Cancer Institute in the Era of Precision Medicine

Charles A. Kunos, C. Norman Coleman

International Journal of Radiation Oncology Biology Physics, 2018

I have always wondered about the processes that fuel the current research programs. In fact, its fascinating to understand and learn how the various processes work in a different cultural context. Herein, I present an editorial recently published in the Red Journal (2018) and have highlighted the excerpts of what I feel are the most pertinent issues.

Current NCI Cancer Therapy Evaluation Program (CTEP) initiatives for radiation therapy devote resources and study to novel combination radiation plus agent study. Here, CTEP defines a radiation plus agent study as any-type radiation therapy given in close proximity (same line of treatment [first-, second-,etc]) before, during, or after therapeutic drugs or biologic agents

It is obvious that bulk of funding goes in identification of “targets”. They are important but not the only goal. Therefore, I am seeing bulk of money going in “biological agents”.

By investigating radiation plus agent combinations at the outset, there is an opportunity to broaden clinical utility.

While this sounds good in theory, the problem with the targeted approach is that we have no idea of how the cancer mutagenesis takes place. Radiation Therapy may also alter the tumour micro-environment and the biological agents may or may not be “effective” in this scenario. If you may remember the famously infamous Bonner trial for Cetuximab and XRT; the benefit was most with the conventional radiation using “field-in-field” (almost analogous to the SIB of today) (a type of altered fractionation) and follow up trials haven’t really shown the cetuximab to be “stellar”. I am not getting in the specifics (its a different sub-site altogether) but these issues have also rankled me.

Overview of the XRT combination

Proposed areas of research in radiation oncology

This is an important pertinent issue:

For example, it is suggested that up to 10 logs of tumor cell kill are needed to sterilize a single targeted tumor, which is a level of cytotoxicity possibly attained by radiation therapy alone, but better achieved when an agent is co-administered during radiation therapy

Bulk of what we know is primarily empiricism. The authors have also highlighted this aspect. For example, the XRT combination with Cisplatin, wherein the earlier trials were motivated by see “whatever-sticks-to-the-wall” approach. We as a community have wisened over the years and now it has been proposed that the newer agents “prove” their efficacy in the “cell lines”

CTEP endorses a preclinical approach that analyzes and emphasizes relevant cell lines (at least 2) and then cell-derived or patient derived (preferably) xenografts orgenetically engineered mouse models of cancer

My beef with it is the issue related to “in-vivo” versus the “in-vitro” model. Like really? Nope, the real time conditions faced inside the cellular environment can never be replicated outside in the petri-dish. But the “hope” is it would be safe (in animal models) with proven efficacy in the cell lines. What do we get in return? Progression Free Survival? Is that the end goal for a hopeless scenario? Is that progress and innovation in cancer?

Further, luckily this article does mention pushing for altered fractionation- hypo fractionation here. At least, there’s an awareness that it might be something better than the “conventional” methodology. I strongly feel that these are motivated by primary concerns of “finishing” off radiation therapy than exploring the true benefit of giving large doses per fraction. I am reminded of a beautiful issue of Seminars in Radiation Oncology about the fractionation and whether the venerated LQ model is applicable to giving large doses per fraction. Hopefully, with the trails that are being encouraged to explore the fractionation schedules, we might have a better clarity about the effect of radiation fractionation on tumour micro-enviroment. I remain hopeful about a better understanding on this important pertinent issue.

Reasonable number to initiate clinical trials, although the larger the enhancement the better, especially when considering hypofractionation, because there are few fractions to enhance compared with standard fractionation (2 Gy per day)

A lot of funding is also being made available for immunotherapy and in particular, abscopal effect, if any. This remains an exciting area of research and hopefully, we might have better insight in this. Is XRT having “only-local” effects or does it extend beyond the narrow confines of planning target volume? Interesting!

A proof-of-principle trial utilized local radiation therapy and granulocyte-macrophage colony-stimulating factor to potentially induce an abscopal response among treatment refractory solid tumor cancer patient. A trial evaluating high versus low radiation dose and immunotherapy is underway (NCT02888743) (emphasis mine)

Ah, the heavy particles. This opens another can of worms, isn’t it? Protons/Carbon Ions versus the Photons (traditional). Meta-analysis hasn’t been kind on its use but it probably represents an arms race to push for the lingering effect of higher OER/presumed benefit versus lower rate of side effects. I haven’t been exposed to the working but it remains subject of much interest. The countries listed here are indeed in a technological arms race. My only interest would be in radio-biology; more than the Bragg peak 🙂

The heavier particles, such as carbon ions, already in clinical use in Japan, Germany, Austria, Italy, and China, might offer further radiobiological advantages beyond protons

Radiogenomics offers the best insight. This is closest to the “personalised medicine” that we are talking about and is perhaps the sum combination of what all I have discussed above. I strongly feel that in sensitive patients; i.e. the ones have the driver mutations, there ought to be an “intensification” of treatment with a higher chance of more prolonged remission/local control, rather than the de-escalation of treatment schedules (as is currently in HPV positive oropharyngeal trials). Nope, we want to hit it hard when it matters the most and not otherwise. I think in the entire editorial, this is the most pertinent issue that’s been highlighted.

Single-arm design umbrella trial that aims to test whether patients with a specific tumor mutation, amplification, or translocation of genes in a driver molecular pathway derive clinical benefit if treated by radiation therapy or by radiation-agent combinations specifically targeting the driver molecular pathway. Such a study is akin to the NCI Molecular Analysis for Therapy Choice trial (NCT02465060) (emphasis mine).

Further aside in the ongoing discussion about the “personalised medicine”:

New radiation therapy trials are taking a first step to identify tumor mutation, amplification, or translocation of genes that drive radiation therapy sensitivity or resistance (NCT02888743,NCT01096368) (emphasis mine) 

The bureaucratic nightmare! The process of drug discovery and final clinical trials!

Last but not the least, the authors have not forgotten about the most important subset of patients: kids! It breaks my heart to see them in the hospital but someone has to care for them! Its important though to keep the fundamentals in place here. Delayed developmental milestones are a huge problem because of the human opportunity cost as well as the burden on the healthcare where costs are spiralling out of control. Research, here, would need to identify where paediatric patients need to be treated and how much treatment regimes need to be identified. With a whole different universe of molecular mutations, this area is ripe for disruption.

Whether radiation therapy could be delayed until developmental milestones are reached or whether radiation therapy could be omitted altogether remain important clinical questions for the radiation oncology field. Improving pediatric patient selection for radiation therapy, perhaps through molecular prognostic factors, is a good example of future radiation therapy science better anticipating needs for intensive local therapy or for intensive systemic therapy.

Balance between “evidence” and “trials”

I was alerted to this in my twitter stream today- what is the importance of the randomised trials versus the “real world evidence”. The authors have mentioned about the high cost of “randomised trials”- which of course, are necessary in order to define compliance with the regulations.

In fact, I have always felt that the drug trials are “too good to be true”; and overwhelmingly “positive”; especially if they are biologics. Ipso facto, you can’t argue with the logic of blocking pathways. How efficacious are these is left to open judgement and by vanity metrics like “progression free survival”. Albeit, the cost to define the overall survival is very expensive, no doubt.

There’s a new kid on the block- Real World Evidence (RWE). From the quoted write up:

Although the definition of RWE is evolving, most associate RWE with data derived from medical practice among heterogeneous sets of patients in real-life settings, such as insurance claims data and clinical data from electronic health records.

My only contention- how do they address the heterogeneity in “real world data”? How will they weed out any erroneous assumptions? How will they actually separate the manifestations of other competing co-morbidities (if any) with the presumed actual effect of the drug in isolation.

RWE provides important insights into patterns of care, limitations to market uptake, health care use costs, and discovery of toxicities otherwise masked in highly selected patients inherently enrolled in RCTs.

This point is perfectly valid. The selection of patients in trials is subject to a lot of bias at the outset. Indeed, its a foregone conclusion that recurrent/advanced/progressive patients are unlikely to have “improved” outcomes- save the firepower for the “preserved” patients. But that skews the results towards a more favourable subset.

I am more interested to know how the actual molecular profiling will help in patient selection; especially for definitive treatments like radiation therapy. “Precision Oncology” is after all personalisation of treatments. More importantly, it should help in pre-selection of patients for dose escalation without the increased incidence of side effects. That would be the holy grail of treatment schedules.

A great read and food for thought.