New Ebola Drug 100% Effective In Monkeys

TrisexualPuppy writes "A team of scientists at Boston University has created a cure for the Ebola virus, first discovered in 1976. After setting the correct dosages, all monkeys tested with the vaccine survived with only mild effects. No tests have been performed on humans yet, as outbreaks happen infrequently and are difficult to track. Quoting NPR: '[The drug] contains snippets of RNA derived from three of the virus's seven genes. That "payload" is packaged in protective packets of nucleic acid and fat molecules. These little stealth missiles attach to the Ebola virus's replication machinery, "silencing" the genes from which they were derived. That prevents the virus from making more viruses.'"

Comments

Re:This brings to mind...

[nashv]

The method with which the "drug" works is called RNA intereference. RNAi is more or less a standard method in molecular laboratories. Unfortunately, the efficacy of RNAi in different cells and for different proteins varies a lot, for reasons that are poorly understood. Further, RNA is rather unstable in water, and delivering substantial doses of RNA to cells in an organism has remainded challenging.

Morever, all viruses do not start with an RNA-based genome. Some DNA based viruses use promoters for their genes that cause very strong expression of the genes, like the CMV promoter [PDF alert], which is used in isolation to create "over expression" in molecular biology. RNAi is typically very poor against such strong promoters.

Ebola is a virus that is relatively slow replicating in the initial stages. It is not a particularly ingenious design as compared to say the flu virus. This gives the RNAi a chance to work against it.

In short, I don't want to say _never_ (that'll just be ignorant), but as yet, RNAi needs a lot of research and is perhaps not the best strategy for all viruses.

Re:100% effective in FIVE monkeys

[Daniel Dvorkin]

The p-value is 0.00032 by my off-the-cuff calculation (pbinom(0, 5, 0.8) in R.) So yeah, it's pretty significant. That being said, sample sizes this small still do tend to make people nervous -- the p-value is calculated assuming that the monkeys in question represent a good sample of the population, and doesn't account for lab-specific or family-specific effects. (Where were the monkeys bred? How closely are they related? What sub-population do they belong to? Etc.) So we can certainly accept the finding for what it is, but regulatory bodies will, with good reason, want to see larger animal trials before approving even limited human use.