Trillium Therapeutics Inc (TRIL)

[NY1972]

A 'rebuttal' from Dewophile:

Thx I hadn't read this interaction before. From jb 118:

Quote:
I agree with Peter and others that the DLT at a low dose itself shouldn't be too worrisome, although it seems like it was a bit of a surprise given that they might have gone 50x as high if I remember the trial design.

Although there will be more data presented than the abstract, I don't expect to come away with a substantially different impression than we have today. I think that will take more time to play out


then Miljenko:
Quote:

Yes, I know that is trade-off between macrophages and platelet, but when company need to reduce dose indicate that toxicology/preclinical were not properly and completely done??? Their resource ($$$) were limited, and if they need PIPE certain found will short shit out of it as long as they smell profit


It's clear that platelets, like RBCs, are physiologically cleared via CD47 and their drug does bind human platelets, so if one were to guess based on animal data the inference would be that DLT in humans would be thrombocytopenia (or leukopenia, which was also prominent in monkeys). Now did the particular dose at which this occurred catch the company off guard? That I don't know. If so, the smart move would be to do further dosing work. gradual titration, pretreatment with promacta, etc., unless there is good reason to believe the current dose should have efficacy. The company addresses this in the abstract, and I would think this is a point of further discussion monday.

as for preclinical tox - they did in fact give some monkeys much higher dose, and the only SE they saw was in hematopoeitic cells. That is why they went with the Ig1 rather than ig4. So despite CD47 being expressed fairly ubiquitously, only hemeatopoetic cells were affected. One can hypothesize all they want about balance between pro and anti phagocytic signals, but at the end of the day the biology of CD47 that we know best is that it is primarily to remove circulating cells targeted for destruction, and probably that is why normal tissues aside from heme are spared. I also think evidence that cancers co-opt this system is strong. So even if the drug can hit circulating cancer cells that could limit metastasis, that alone could be powerful as an added agent to a drug that can hit the primary tumor. I would think the potential is appealing enough for a large pharma to take a flyer on this - especially give how crowded the IO space has become. Something that can set you apart from the competition can have massive payoff. It also seems to me the competition is weak. we don't know much about CELG,but shouldn't 47inc be worried their drug didn't affect platelets, a cell with high CD47 receptor density and which is physiologically known to use CD47 signaling for clearance (and therefore have the requisite phagocytic signals)?
One last comment about dosing. TRIL has an antigen sink problem as well - platelets and WBCs. There are about 10x more circulating RBCs in blood vs platelets/mL, and about 100x more per ml than WBCs. CELG and surface oncology may not cause hemagglutination but they probably do still bind RBCs. That is probably why CELG went with an Ig4 too (so i have heard). So when comparing dosing not only is the TRIL drug much more potent (Ig1 vs 4), it doesn't have nearly the sink that 47 and others are contending with. I bother to mention this bc I think in part TRIL weakness is due to perceived lack of efficacy from their competitors, but there are enough distinguishing factors between the 2 approaches that this can be explained away to some degree
JMO

http://investorshub.advfn.com/boards/read_msg.aspx?message_id=126987783