There has been an interesting exchange of letters in the correspondence columns of Blood (Blood 2010; 116:1187-8) between Ken Foon and Michael Hallek concerning the end-point of treatment. Confusion arises because what was published in the Guidelines in Blood in June 2008 was not agreed by all the authors and it had to be revised in the electronic records in December 2008.
The problem arose because the 1996 Guidelines were widely used for both clinical practice and for clinical trials, and these suggested that a complete response should mean that there were fewer than 30% lymphocytes in a bone marrow trephine following treatment. If the trephine showed lymphoid nodules then the type of remission should be categorized nodular PR. Since 1996 there has been evidence that a deeper remission than just <30% lymphoid cells in the bone marrow, might be more advantageous to the patient, and also that it is possible to distinguish whether a nodular PR comprises tumor tissue of just reactive lymphocytes.
In order to retain a comparison with older trials, the new Guidelines now suggest that the old (1996) criteria are retained, but that in trials aimed at maximizing CR rate, assessment should be made of minimal residual disease by 4-color flow and that lymphoid nodules should be assessed by immmunohistochemistry. Why should this not be done for all patients? I guess it is because the experts are not sufficiently convinced that attempts to wipe out all trace of discernible CLL is beneficial to the patient - witness the recent CALGB trial of post-treatment Campath.
The second point concerns the use of CT scans. The Guidelines suggest that CT scans are not required for either the initial valuation of follow-up in clinical practice. However, they recommended that they be used in clinical trials - one at the start of therapy and the other at first restaging after therapy if it had been previously abnormal. A second CT is not necessary if residual disease can be detected some other way - such as blood tests or clinical examination.
I think this is now clear.