Dose-effect relationship for cataract induction after single-dose total body irradiation and bone marrow transplantation for acute leukemia.

PURPOSE

To determine a dose-effect relationship for cataract induction, the tissue-specific parameter, alpha/beta, and the rate of repair of sublethal damage, mu value, in the linear-quadratic formula have to be known. To obtain these parameters for the human eye lens, a large series of patients treated with different doses and dose rates is required. The data of patients with acute leukemia treated with single-dose total body irradiation (STBI) and bone marrow transplantation (BMT) collected by the European Group for Blood and Marrow Transplantation were analyzed.

METHODS AND MATERIALS

The data of 495 patients who underwent BMT for acute leukemia, who had STBI as part of their conditioning regimen, were analyzed using the linear-quadratic concept. The end point was the incidence of cataract formation after BMT. Of the analyzed patients, 175 were registered as having cataracts. Biologic effective doses (BEDs) for different sets of values for alpha/beta and mu were calculated for each patient. With Cox regression analysis, using the overall chi-square test as the parameter evaluating the goodness of fit, alpha/beta and mu values were found. Risk factors for cataract induction were the BED of the applied TBI regimen, allogeneic BMT, steroid therapy for >14 weeks, and heparin administration. To avoid the influence of steroid therapy and heparin on cataract induction, patients who received steroid or heparin treatment were excluded, leaving only the BED as a risk factor. Next, the most likely set of alpha/beta and mu values was obtained. With this set, the cataract-free survival rates were calculated for specific BED intervals, according to the Kaplan-Meier method. From these calculations, cataract incidences were obtained as function of the BED at 120 months after STBI.

RESULTS

The use of BED instead of the TBI dose enabled the incidence of cataract formation to be predicted in a reasonably consistent way. With Cox regression analysis for all STBI data, a maximal chi-square value was obtained for alpha/beta = 1.75 Gy and mu = 0.75 h(-1). When Cox regression analysis was applied for patients who had no steroid treatment after BMT, a maximal chi-square value was obtained for alpha/beta = 1 Gy and mu = 0.6 h(-1). Cox regression analysis was repeated using the data of patients who had not received posttransplant steroid treatment and also no heparin administration; we found alpha/beta = 0.75 Gy and mu= 0.65 h(-1). An increased cataract incidence was observed after steroid treatment of >14 weeks and heparin administration.

CONCLUSION

The alpha/beta value of 0.75 Gy and mu value of 0.65 h(-1) found for the eye lens are characteristic for late-responding tissues. The incidence of cataract formation can now be quantified, taking into account the values calculated for alpha/beta and mu, TBI dose, and dose rate. Also, the reduction in cataract incidence as a result of lens dose reduction by eye shielding can be estimated.