Roland, you need to investigate claims further before you promote them. People in medicine have been trying to deliver specific radioactive isotopes and monoclonal antibodies to tumors for a decade or more. The problem is that tumors usually do not have precise surface molecules that identify them as a cancer. If they did then the immune system would eliminate them. Is there a subset of tumors with common molecules that can be attacked? Most probably yes — just this week gefitinib/Iressa(TM) was shown to be effective against non-small cell lung cancer. But it only works in 10% of lung cancer cases!

The approach that Dr. Halas has may be reasonable for breast cancer or skin cancer. But it is highly questionable whether it would be useful for liver or brain cancer (where light or IR are unlikely to be able to penetrate). In addition there is nothing in the MIT Tech Review article that explains why nanoshells kill only cancer cells. It does not suggest that nanoshells prefer to target cancer cells (and why they would do so) and as a result one should expect the same results from the use of nanoshells that one gets with chemotherapy (healthy tissue damage in the process of killing cancer cells).