First Anti-Cancer Nanoparticle Trial On Humans a Success 260
An anonymous reader writes "Nanoparticles have been able to disable cancerous cells in living human bodies for the first time. The results are perfect so far, killing tumors with no side effects whatsoever. Mark Davis, project leader at CalTech, says that 'it sneaks in, evades the immune system, delivers the siRNA, and the disassembled components exit out.' Truly amazing."
Re:Too small a sample size (Score:5, Informative)
Well, what's meaningful is that they all didn't up and die, and that a bigger round of testing is to go forward.
Re:Too small a sample size (Score:4, Informative)
The point of the study is to make sure that people don't explode when the procedure is performed, or for something similarly unpleasant to happen--it's a Phase I study, not a real effectiveness trial.
Not just cancer! (Score:5, Informative)
From comments on TFA, "The Lab" writes: "a science editor would be more capable of pointing out what is really exciting here, which is the ability to stop cells from producing a given protein."
I think the cancer aspect is great (if it works) but this has potential for curing a whole host of diseases.
Now we just need to figure out how to change people's DNA on the fly.
Then take a statistics class (Score:4, Informative)
As long as the subjects have the same distribution as the population, this sample can be considered representative of the population. This means that they didn't pick 15 terminal patients and didn't pick 15 100%-survival-rate patients. You can achieve quite a lot when your sample is well selected.
Re:Targetting (Score:5, Informative)
Re:Nice if true (Score:5, Informative)
Re:SWEET SUCCESS (Score:3, Informative)
Re:Targetting (Score:5, Informative)
Re:Artificial virus (Score:2, Informative)
So, they made an artificial virus...
Fail.
Re:Targetting (Score:4, Informative)
I see your point that this does not seem like the ordinary concept of targeting. But in pharmacology that is exactly what targeting is.
Re:Too small a sample size (Score:5, Informative)
I cannot see anything meaningful coming from such a small sample size. It has potential but obviously much more research is needed.
You can't just jump from rats to tens of thousands of humans. That's why the sample size is 15. That's why it's a Phase I trial. There are four phases of clinical pharmaceutical testing that follow preclinical (animals, in vitro, etc.) testing. Phase I normally tests a treatment in healthy humans in order to see the negative effects of the treatment (this is not necessarily the case in cancer treatments because all cancer treatments have significant negative effects). Phase I trials are only a couple dozen people, max. Successful Phase I trials allow for Phase II trials. These usually have one or two hundred people with the disease the therapy is intended to treat. In Phase II, they are mainly gathering pharmacokinetic data (half life, metabolism, volume of distribution, etc.). Phase III is where you start to see the trials you're clamoring for. These are typically done in several thousand patients, all with the disease in question. These trials are placebo-controlled, randomized, double-blind studies (the hallmark of research). Statistical analysis then allows you to determine if the therapy was effective in improving outcomes. If so, the drug goes to the FDA. 30 days later, it is officially on the market. Phase IV studies begin here, and continue perpetually. They are called post-marketing surveillance, and they study long-term effects (because previous trials are not long enough to do this), as well as very rare adverse effects (where the sample size in previous trials may have been too small to correctly detect the progressive multifocal leukoencephalopathy that occurs in 0.1% of patients treated).
So don't claim the study size wasn't big enough - it wasn't supposed to be. Phase III trials are what you want. Phase I and II trials are of no interest to anyone outside of health professions, really.
Re:Targetting (Score:5, Informative)
How else would you define targeting in this context other than to mean only binding to cancerous cells? It seems you are implying that targeting can only refer to conscious 'aiming', but that is only a subset of things that can be considered targeted.
Targeted can mean 'select as an object of attention or attack'. That is what they are doing when the design a drug.. selecting cancer cells for attack, and then designing the drug so it will only effect those cells. Check out http://en.wikipedia.org/wiki/Drug_design [wikipedia.org]
Targeted drugs DO mean something specific in pharmacology.
Re:Isn't this... (Score:2, Informative)
The first? Hardly... (Score:3, Informative)
.
Abraxis has been around for, literally, years.
Re:Targetting (Score:3, Informative)
That's because of a stupid law... (Score:5, Informative)
> everything is known to cause cancer in California... I could never figure it out, so I just stay away from California.
Everything says it causes cancer because of Proposition 65 [wikipedia.org]. Basically, if something in California is known to cause cancer (even only if ingested by the ton), you have to label it, or lawyers can sue you under a "private attorney general" law. In theory it might be a good idea, but it was implemented so that the defendant has the burden of showing that it's basically impossible to the nth degree that the thing could cause cancer in the quantities you're talking about.
This resulted in a lot of litigation where basically lawyers went around everywhere and said "Oh! You have flame-retardant furniture! Did you know it can cause cancer if you lick it?" "You're a dentist! You use drugs that can cause cancer if you administer them for a week and you didn't post a notice!"
This resulted in a plethora of notices to prevent lawsuits--notices which the public ignores because they're on everything. So in the cases where the warning is actually important, it gets ignored because there are so many.
IIRC, there have been some efforts by the AG (and some courts) to limit abuse.
It is spelled Caltech NOT CalTech (Score:3, Informative)
http://media.caltech.edu/press_releases/13334 [caltech.edu]
If you cannot spell Caltech properly - please turn in your nerd card.
Re:Targetting (Score:5, Informative)
Well, the summary says that the results have been perfect so far, so that's a real good sign. Frankly something like this sounds too good to be true, but every once and a while breakthroughs do happen, so we can hope!
Re:Then take a statistics class (Score:3, Informative)
Self reply ...
I just feel I should clarify, in order to get the perfect distribution you need a sample size at least big enough to all possible variations in the distribution in the ratios present in the population.
I.E. The samples too small to provide useful statistical results. Its just a general indicator.
If all 15 people live perfectly for a normal life after this it doesn't mean its 100% safe, it just means its highly likely to work well on a lot of people, but its entirely possible that those 15 people happened to share something that no one else has. Likewise if all 15 people die tomorrow, its a really good indication that this isn't going to be that useful on the population, but it doesn't mean that everyone will die from it ... those 15 people may just have been exposed to a secondary compound since they are all early test subjects that happened to interact and kill them.
The sample size is just too small to provide truely useful stats when it represents about 4 bits of diversity in a population that is defined by a code that allows for 38 bits of variation currently representing about 33.5 bits of that.
Its funny how engineers like to over sample. An engineer wouldn't accept a sample size less than 39 bits to be safe, even though the total population is smaller than that currently. Statisticians on the other hand are happy with pretty much any number greater than 1 bit.
Re:Too small a sample size (Score:3, Informative)
Re:Targetting (Score:5, Informative)
Why is this the first example of the concentration of nanoparticles showing a correlation with the initial dose? To me it seems obvious that this would happen, so I'm curious as to what normally prevents this.
The dose-response characteristics of a substance carries important information about its pharmacokinetics, such as how rapidly it is metabolized and excreted. It is NOT obvious that increasing dose always results in increased concentration in the targeted tissue(s), just as it is the case in chemical reactions in which increasing the proportion of one reagent may not necessarily increase the yield. For example, increasing dosage may also increase hepatic clearance. What the researchers demonstrated basically boils down to this: the nanoparticles' activity is highly specific to the tumor cells and does not get "broken down" or excreted proportionally higher as the dose increases. Such characteristics are highly desirable.
Re:Targetting (Score:5, Informative)
It would be entirely possible to target the RNA sequence to only bind to malignant cells and ignore normal ones.
Yeah but chemotherapy and radiotherapy work the same way. The problem is that the characteristic of cancerous cells they bind to is the fact that they grow fast.
No they don't work the same way. Radiotherapy, and most current gen chemotheraputics, work against all dividing cells. It sounds like these nanoparticles use a specific protein (NOT rna) to bind to the cancer cell, then once inside they cause RNAi (this is where the RNA comes in) to knock down a specific gene transcript.
I obviously don't know the specifics, but if you make a nanoparticle that binds to and is taken up by cells expressing a specific growth factor, that's -not- going to be taken up by all cells. There are a lot of growth factors, and often in cancers, growth factor receptors are expressed far more than they normally are. So the targets are found only on some cell types, and there are hundreds more on the cancer cells than the healthy cells.
Furthermore, you could target individual genes to be knocked down by the RNAi effect, potentially genes that aren't even expressed by the normal cells which are expressing the receptor you're targeting. I'd guess for maximal efficiency, you'd be targeting housekeeping genes that all cells needed, but in principle you could make it a gene that cancer specifically needed.
Both levels of specificity, even if they're not used, are a far cry from "damage every dividing cell and hope you kill the cancer before you kill the dividing tissues the patient needs to survive." And they don't actually use RNA to bind to the cell, they use protein to target the cell and RNA once inside the cell to target the specific gene. ... by the way, I am not a molecular cancer biologist.
Re:This was already posted on /. 2 days ago! (Score:2, Informative)
And that time they spelled Caltech correctly!
Re:Targetting (Score:3, Informative)
However, yes, it is like chemotherapy in that it will affect proliferating healthy cells as well as cancer cells (could be a lot less toxic in terms of metabolic clearance compared to current oncolytics). The protein they cut production on using RNA interference, ribonucleotide reductase, is already targeted by a few chemo drugs in use like gemcitabine. Any cell that needs to make DNA bases needs a working copy of ribonucleotide reductase; rapidly dividing cancer cells just have a much more urgent need. However, this nanoparticle/siRNA system has the advantage of being highly adaptable. Find a new oncogene target? Write a new siRNA to silence it, and you have the equivalent of a whole new class of cancer drug.
Re:Artificial virus (Score:4, Informative)
I didn't see any indication that the nano-particles are self-replicating, or capable of spreading from one person to another, so you'd need to inject each target individually. It's probably easier just to shoot them.
Plus, if I understand correctly, cultural conceptions of race don't map very well to genetic differences. So finding a race-specific gene to target might be harder than you'd think.