Gbm and contralateral hippocampus
Started by Ray D. Ayshun
We contour the hippocampus (left, right, and together; with no expansion) for all VMAT cases and use it as an avoidance structure. It takes a minute to put it on if used to doing it. Generally we don't put hard constraints- but prefer to see lower mean doses.
If not going to affect your coverage, then why not?
@Ray D. Ayshun - great question! I'm interested in learning what others with more experience do, but can share my thought process thus far:
I tend to contour C/L hippocampus and aim for ALARA, w/o sacrificing coverage as @evilbooyaa said
Old Mednet post I read few years ago by Vinai Gondi & Minesh Mehta mentioned hippocampus D100% ≤ 10 Gy and Maximum dose ≤ 17 Gy
Some interesting papers I've read related to this topic
1. Pretty good viewpoint on this subject: Hippocampal Avoidance for Gliomas
2. Less Hippocampus Atrophy with <10 Gy mean dose: Radiation dose-dependent hippocampal atrophy detected with longitudinal volumetric MRI
3. Primary brain tumor EQD2 to 40% of B/L hippocampi >7.3 Gy related to neurocognitive impairment: Hippocampal dosimetry predicts neurocognitive function impairment after fractionated stereotactic radiotherapy for benign or low-grade adult brain tumors - PubMed
4. RT dose & effect by brain substructure: https://www.thegreenjournal.com/article/S0167-8140(20)30154-7/abstract
I tend to contour C/L hippocampus and aim for ALARA, w/o sacrificing coverage as @evilbooyaa said
Old Mednet post I read few years ago by Vinai Gondi & Minesh Mehta mentioned hippocampus D100% ≤ 10 Gy and Maximum dose ≤ 17 Gy
Some interesting papers I've read related to this topic
1. Pretty good viewpoint on this subject: Hippocampal Avoidance for Gliomas
2. Less Hippocampus Atrophy with <10 Gy mean dose: Radiation dose-dependent hippocampal atrophy detected with longitudinal volumetric MRI
3. Primary brain tumor EQD2 to 40% of B/L hippocampi >7.3 Gy related to neurocognitive impairment: Hippocampal dosimetry predicts neurocognitive function impairment after fractionated stereotactic radiotherapy for benign or low-grade adult brain tumors - PubMed
4. RT dose & effect by brain substructure: https://www.thegreenjournal.com/article/S0167-8140(20)30154-7/abstract
Cool, thanks. Those were, more or less, the constraints I was gonna go with, D100 < 9 Gy and dmax 16 as per HA-WBRT. thanks for the links.
I haven't heard that but it makes sense. I'd imaging the same thing is true for other brain substructures, or for other malignancies/organ subsites with disease outside of the primary). Breast is one example.
However I can state with much certainty that I'm not going to electively treat the hippocampi for gliomas anytime in the near future.
High dose vs low dose irradiation of the subventricular zone in patients with glioblastoma—a systematic review and meta-analysis - PMC
The published data indicate that the irradiation of the subventricular zone (SVZ) might play a role in the treatment of patients with glioblastoma (GBM). We aimed to determine whether radiation treatment doses (high vs low) applied to the SVZ can ...
www.ncbi.nlm.nih.gov
I would only send these patients to a high volume palliative fellowship trained rad onc.
Advertisement - Members don't see this ad
Well, I'm gonna spare the hippocampus. I'll let y'all know if there's a bad outcome...
Gbm is Palliative? I guess that's not a totally unreasonable perspective...
I spare it. It is usually easy as GBM usually lateralized. If you don't explicitly spare it, typical field arrangements lead to pretty good dose there (well above above mentioned contraints) routinely. May be useless but cost is close to nothing.
Oh I send all my patients only to a high volume palliative fellowship trained rad onc.
Somehow the hippocampus became an idée fixe in rad onc. Completely unbuttressed by any good, compelling human radiotherapeutic data that I know of, people focused on rat brains theorized that ~3/4ths less XRT dose to a human hippocampus would measurably improve outcomes vs regular WBRT. This theory's on the ropes. Which is fine. Sometimes things just don't work out. We need to let the hippocampal obsession go (and I won't even complain if Evicore won't pay for it in WBRT). In GBM, one way to spare the contralateral hippocampus is don't put a huge old-school margin on your volume, and do as conformal a plan as possible. It's like all the benefit with none of the busy work of hippocampal contouring.
Well nobody's doing trials to determine how irradiating extremities in sts affects cognition. Anatomically, the brain seems important for cognition. Seems reasonable to think there's a most important part of the brain when it comes to that. I'm voting against the motor strip.
I definitely get it. But it's tough to be strident about what brain parts are or aren't important in humans.
yes, always contralateral sparred where i was trained. Protons to spare if needed!
Oh the old “theres this rat paper” radiation thing. Humans are just large rodents
Just remember it's a GBM
Advertisement - Members don't see this ad
Agree with all of this and it's what I do when I can. If you can't meet it, I don't stress. I don't edit PTVs out of it or lower PTV coverage or anything like that.
The problem with this scenario is that all the data ends up confounded because the more central or bilateral tumors do worse anyway, and progressing tumor causes a lot of neurocognitive dysfunction.
Now is contralateral sparing a good justification for protons in primary GBM? I know some docs who would argue this to be the case! Some patients eat it up too--there's this belief out there that protons must be better for everything since they're more expensive and limited to a smaller number of institutions.
Ew. Protons for GBM is like one step away from protons for palliation. C'mon man!
2 steps away from G6 prostate ca, but whatever pays the bills....
Wouldn’t you want the very best most beautiful plan? Don’t lie
Need to reduce “low dose bath” in case got to come in from the front for a lung met, leaves more room
I've been doing this more often with my definitive CNS cases. There is no hard constraint that I am aware of at this time.
