DAT ± AHA = bad CLL

The benefits of public transportation may not always be apparent. Robin Coombs conceived the basis of the direct antiglobulin test (DAT) while traveling back to Cambridge on a slow, blacked-out train during the Second World War.¹  Armed with this test, it became apparent that a positive DAT and overt autoimmune hemolytic anemia (AHA) occurred in many cases of chronic lymphocytic leukemia (CLL).²  Subsequently, AHA was shown to be more common in advanced than in stable disease, and could be evoked by either drugs or irradiation. AHA following fludarabine can often be difficult, sometimes impossible, to control. Autoimmune complications can therefore pose a significant problem in the management of CLL.

Since fludarabine-based combination che-motherapy has now become the standard of care for most patients with CLL, the observations in this issue of Blood by Dearden and colleagues, along with existing results from the UK Leukaemia Research CLL4 trial,³  on the incidence and significance of a positive DAT and the development of AHA in 777 previously untreated patients are of considerable clinical interest. In this trial, patients were randomized to receive either single-agent chlorambucil or fludarabine, or fludarabine and cyclophosphamide (FC) in combination. By every criterion, FC appeared to be superior. The lowest incidence of AHA development during treatment was with FC (5% of patients), compared with 11% receiving chlorambucil and 9% receiving fludarabine. The number of DAT-positive patients with-out AHA fell with FC, but in contrast, rose with single-agent fludarabine. It is particularly noteworthy that the severity of AHA appeared to be worst in patients receiving fludarabine alone; all 4 deaths from AHA occurred in such patients. All things considered, there now appears to be little justification for using fludarabine monotherapy in CLL. Development of AHA was associated with DAT positivity, clinical stage, treatment, age, and serum β2-microglobulin levels. Five-year overall survival was adversely affected by AHA (58% vs 37%), which may reflect that these patients received less chemotherapy. However, it is fascinating that DAT positivity itself, before treatment and in the absence of AHA, was also an independent poor prognostic marker in terms of both progression-free and overall survival. Why this should be remains unknown. Both DAT positivity and AHA are therefore markers of poor-risk CLL.

How such autoimmune complications of CLL arise, in a disease otherwise characterized by progressive immunoparesis, remains uncertain. AHA in CLL usually represents the development of high-affinity IgG anti–red-cell autoantibodies by residual normal B cells. How these develop is not clear, but presumably such development represents a failure of normal T-cell regulatory mechanisms that accompany progressive CLL and its treatment. It will be interesting to assess the development of AHA in relation to IGHV mutational status; these data will be available from the CLL4 trial. Whether use of rituximab during induction chemotherapy will alter the frequency and severity of autoimmune complications in CLL remains to be determined.

We can anticipate more interesting clinical data to emerge shortly from this randomized trial.