From what I understand, the drugs were just less effective than originally predicted. The NHS had an agreement with the drug companies that if the drugs did not perform as well as predicted, then they would not pay the entire cost. Once the study came back that the drugs were less effective, NHS should have paid less, but they continued paying full price.
The article doesn't say the drugs are not effective, they are just not as effective as predicted.
another source: http://www.eurekalert.org/pub_releases/ ... 060310.php
Edit: The financial times paints a bleaker picture:
http://www.ft.com/cms/s/0/f37adede-6f2e ... abdc0.html
Some 5,000 patients were recruited. But far from slowing the disease, progression was faster in the patients who had the drugs than in untreated patients – although the comparison was with a large, past group of untreated Canadian patients.
The results were so poor that “the manufacturers would need to pay the NHS to use the drugs to make them cost effective”, Christopher McCabe, professor of health economics at Leeds university, says in the British Medical Journal.