[sam537]

Oncologist here. You can think of drug development as an extremely wide funnel with a minuscule exit hole. At any given time a medium sized pharmaceutical company has 300-500 drug candidates. From this group 50 make it to phase 1 trials. 40 get slashed due to toxicity, company abandons them, gets bought by third party and sits in sleepy storage. 10 advance to phase 2 trials where 5-8 may not end up having enough efficacy/too toxic, does not improve survival. The remaining 3-5 advance to phase 3 where they can suffer the above as well.

Tl:dr: Things usually work in the lab where all ideas start. Until a compound goes through thorough human experimentation believe little.

[dvirsky]

Sounds like this one is pretty advanced down the funnel, having successfully passed phase 1 human trials, or am I missing something?

[abcc8]

Phase 1 trials are used to determine drug toxicity. All that reault really says is that the new drug won't itself kill the patients. It still needs to be compared relative to the standard of care in larger phase 2 and 3 trials.

[tempestn]

However, the article also describes patients in the trial having positive effects. Yes, it's not an efficacy trial, but if it was specifically used on patients who have exhausted all other clinical options, there were no significant adverse side-effects, and there were beneficial effects for some substantial number of patients, that sounds pretty hopeful to me. Of course it's possible issues will still be discovered, but this is a long way from a top-of-funnel lab result.

[nighthawk454]

Thanks, that's an interesting insight into the dropoff rate for these ideas. Do you have an opinion on what the bottleneck is? Volume of ideas, length of testing, or perhaps flawed approach in general?

[sam537]

This is the bottleneck for one company. We have many companies this size doing the same thing.

The issue is large companies (novartis/BMS) sometimes sit on compounds and refuse to develop because they think it may not be worth it, may think a certain disease (read market) is saturated, or they may be waiting for the 'right time'. It can be frustrating.

Smaller companies have 2-3 compounds which they actively work to develop but most of them end up getting acquired by the big fish.

[pfdietz]

The extreme complexity of biology. There's no substitute to just trying things and seeing if they work, because you will never understand what's going on beforehand.

[dominotw]

funny how many of blockebuster drugs were discovered by accident.

[pfdietz]

And for some that were designed to hit a specific target, it was discovered afterwards that they work by a completely different mechanism.

[Gatsky]

Glad to see another oncologist on HN!

[ncmncm]

Thanks. I had read of success infecting tumors with virus, that the immune system will control only outside the tumor. But I didn't hear any more about it. What happened with that?