Two provoking abstracts on CNS prophylaxis at #ASH2020. Credit to the U.S & Canadian authors for attempting to address the thorny issue of CNSp in DLBCL.

However: substantial biases & lack of key data limit interpretation. Unfortunately these data don’t turn the dial....THREAD: https://twitter.com/mtmdphd/status/1336744285035696128
The 2-centre Canadian study IDd ~300 patients who fulfilled criteria for CNSp but only ~100 rcvd HDMTX; this immediately confers bias+++ in any attempted comparison of 2 (different) gps. Thus ~100 pts rcvd HDMTX - a small sample size in this context = HUGE confidence intervals.
Importantly, the authors of the Canadian study did not yet describe what % of the CNS events were isolated to CNS vs synchronous CNS&systemic. The absence of this key information makes interpretation even more difficult.... two v different clinical problems.
The large multi-centre US study also had substantial limitations & inherent biases. The exclusion of pts who rcvd both IT & IV for a cleaner analysis immediately introduces bias. 

Moreover It was not known whether MTX given as a 24h or 3h infusion...which is rather crucial...
Similar to the Canadian study, it was not stated what % of the CNS events were isolated to CNS vs synchronous CNS&systemic. Lastly the characteristics of the 2 groups were not balanced (number of EN sites and testicular)...
CNS prophylaxis in DLBCL remains a v important problem to solve Large carefullly designed retrospective data can be informative. Great to see upcoming Ph3 studies permitting EOT HDMTX. Biomarker data required to inform/refine - GOYA (GCB vs ABC by GEP) gave us a glimpse of this.
You can follow @DrChrisPFox.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled:

By continuing to use the site, you are consenting to the use of cookies as explained in our Cookie Policy to improve your experience.