Precise Prediction of Response to Therapy by Using the Chemosensitivity Index C_i in Patients with Chronic Lymphocytic Leukemia.

There are quite a few number of prognostic factors known to determine the prognosis of patients with chronic lymphocytic leukemia (CLL). However, the individual response to therapy may still not be predicted by these factors. Therefore, we investigated prospectively the prediction of response to therapy and the prognostic relevance of the in-vitro chemosensitivity profile in patients with CLL within a randomized trial. For the evaluation of the chemosensitivity profile we applied the methodology of the chemosensitivity index C_i, which we described before in AML (Staib et al., Br J Haematol 2005). Peripheral blood samples from patients with the established diagnosis of B-CLL were taken with written informed consent before treatment was started. For in vitro chemosensitivity testing the Differential Staining Cytotoxicity (DiSC) Assay was used. The fresh leukemic cells were incubated with fludarabine, bendamustine and six other cytotoxic drugs. DiSC assay results were expressed as percent tumor cell survival (%TCS). The chemosensitivity index C_i was calculated according to dose-response curves for each clinically used drug and the area under the curve (AUC). The cut off point between resistance and sensitivity was adjusted at 0,5 for each drug by the AUC data of a subgroup of clinically resistant patients: C_i ≤0.5 indicates resistance, C_i >0.5 sensitivity to the drug.

Results: 53 patients were recruited to a two-arm randomized second line treatment trial from 2001 to 2005. The patients were treated with either fludarabine or bendamustine as a monotherapy. For all patients a chemosensitivity test was performed to predict the clinical outcome for the used treatment. 25 (47%) patients received fludarabine and 28 (53%) bendamustine. 41 pts. (77%) achieved a complete or a partial remission whereas 12 pts. had a stable or progressive disease (23%). All of the responding patients were predicted as sensitive (100%), and all 12 non-responding patients were predicted as resistant (100%). The over all accuracy was 100% (Pearson Chi-square p<0.001). The progression free survival time in patients who were tested to be resistant was 4 months compared to 22 months in patients who were tested to be sensitive (p<0.001).

Conclusion: The presented study demonstrates that the in-vitro chemosensitvity profile is able to predict the clinical treatment outcome in patients with CLL precisely. Therefore, the in-vitro chemosensitivity testing evaluated by the C_i may serve as a powerful tool for assay directed and individualized therapy strategies in B-CLL.