What has got the CLL world talking at the moment is a paper in Cancer from the Mayo Clinic which suggests that patients treated by the CLL team live almost 2 years longer than those treated by teams at the Mayo that are expert in other hematological conditions like myeloma, lymphoma or MDS. Why should this be?
This was not a randomized prospective trial so the obvious explanation would be that the groups were not well matched. However a multivariate analysis of known prognostic factors such as age, stage, sex, and total white count still leaves the expertise of the doctor as a significant factor.
The most striking difference between the two groups was that the experts kept the patients on watch and wait longer. Once treatment starts then hazards are introduced. The patient is more likely to develop infections, autoimmunity, extra chromosomal abnormalities, pancytopenia, Richter's transformation and MDS. The commonest mistake I see made by general oncologists is to treat just because of a high white count - not an indication in the international guidelines.
Non-experts are much less likely to request biological prognostic markers; so much so that it was not possible to separate the two groups of patients according to these parameters. Even the cheap Beta-2M test was not done sufficiently frequently, so it was more likely ignorance rather than expense that was the deterrent.
Non-experts were much less likely to enter patients into clinical trials even though it is well known that patients entered into trials fare better tan those treated according to physicians' choice.
Non-experts were more likely to choose chlorambucil or rituximab alone than a purine analog combination. Now I think there is a place for chlorambucil alone in the right patient, especially at the right dose, but this should be chosen from a position of expertise rather than ignorance.
So I find the paper believable. It is not necessary to be treated by an expert - their fellows do even better than they do, but make use of an experts expertise.