Is it ever appropriate to omit temozolomide in unmethylated glioblastoma?

Perhaps a different perspective on this question would be which unmethylated patients would you be willing to not treat with up-front temozolomide? The genesis of the question and conundrum comes from the modest benefit described in the above-mentioned trials for this subgroup. As Hegi herself describes, in her opinion, truly unmethylated patients do not derive benefit, but the challenge lies in identifying the truly MGMT unmethylated patients, because most tests have around a 10% or so rate of ...

This is a good question, and of course, the answer hinges upon "why are you considering de-escalating treatment intensity for your individual patient?"
The OS benefit of TMZ in the original Stupp/Hegi report showed an HR of 0.69, p value of 0.06 for unmethylated patients, with median OS 11.8 mo for RT alone and 12.7 mo for TMZ +RT. So it is right to consider this effect statistically marginal, and clinically of marginal value...yet the 2-year rate of survival was 14% for TMZ-RT vs RT alone in unm...

I agree with the detailed reply by Dr. @Dr. First Last.

I would add that in the Hegi NEJM 2005 article, <2% of unmethylated patients treated with RT alone survived 2 years, while 13.8% unmethylated patients treated with RT + TMZ were alive at 2 years. Although this proportion is far less than the 46% of MGMT methylated patients treated with RT+TMZ, it suggests there is a population of unmethylated patients that still benefit substantially. Although, MGMT methylation determinations are better now ...