The Challenge

Being hailed as the biggest breakthrough in biology of the last decade or two and being reproducible in the Petri dish is no guarantee that drugs based on RNAi will successfully make it as drugs. Three main hurdles need to be taken for this to come true: 1) efficient delivery of the RNAi agent to the target cells, 2) managing off-target effects, and 3) endowing RNAi molecules with drug-like properties.

The efficient delivery of the RNAi agent is frequently cited as the main hurdle to the wide application of RNAi. While local delivery such as needle injection into the eye for ocular diseases or inhalation for lung-related conditions has a relatively high likelihood of success, systemic delivery, e.g. needed to reach metastatic cancer cells hidden in the body, is a taller order. However, a number of delivery methods are being tested, some borrowed from older oligonucleotide-based technologies like liposomes, some more innovative. It is likely that no one-fits-all solution will emerge, but strategies that are tailored to the specific disease. Particularly interesting are approaches where a synthetic siRNA is coupled to agents such as monoclonal antibodies or RNA aptamers which can selectively target cells. Hence, the specificity of the siRNA is compounded by the specificity of its delivery. This should also help in reducing the likelihood of potentially harmful off-target effects.

Off-targeting, the suppression of non-targeted genes is mainly a consequence of the siRNA acting like a microRNA. MicroRNAs are related endogenous 20-24 nucleotide small RNAs that recognise their targets through less than perfect complementarity. Although this typically does not downregulate target genes as dramatically as an efficient siRNA might do, nobody can be sure that a 2-3 fold reduction in a “random” gene will not have adverse side-effects. I should stress that this specificity is probably still much better than many other drug-classes today, but we would like to do better, especially when human health and the substantial resources needed for the development of a drug are at stake. Helped by the knowledge of the human genome, bioninformatics already can winnow down the number of potential off-targets, thus reducing the likelihood of an adverse side-effect. More lately, however, and as demonstrated by the work of Dharmacon scientists, it has become possible to modify the siRNA such that it will lose much of its microRNA abilities while retaining potent RNAi-like cleavage potential. I view this as a particular exciting development.

Chemical modification of siRNAs can also help to enhance its drug-like properties such as half-life, metabolism etc. However, it was only quite recently shown quite by Alnylam Pharmaceuticals, viewed by many as the leading RNAi Therapeutics company, that siRNAs may knock down genes with a profile that may allow dosing every 2 to 4 weeks. This is a great relief and opens up RNAi for many more, particularly chronic applications such as hypercholesterolemia, than would have been the case if the knockdown was limited up to 4 days after administration. The latter is typically observed with cultured cancer cell lines and is due to dilution of the siRNAs following frequent cell divisions. For some genetic diseases, viral gene therapy vectors for the expression of hairpin RNAs that are processed to siRNAs by the endogenous RNAi machinery are a further important option. They should allow for the more long-term expression of RNAi effector molecules, although the risk-benefit equation is shifted for DNA-based therapies.

In summary, many strategies are being explored to develop RNAi as a safe and efficient therapy. While human trials for each siRNA will have to be performed to evaluate its specific therapeutic value, as the risk of RNAi Therapeutics gets more manageable each day, it promises to become one of the most specific and versatile drug class to date.