Red marrow radiation dose adjustment using plasma FLT3-L cytokine levels: improved correlations between hematologic toxicity and bone marrow dose for radioimmunotherapy patients.

UNLABELLED: Calculated red marrow absorbed dose in patients receiving radioimmunotherapy (RAIT) has not been highly predictive of the dose-limiting hematologic toxicity observed in many patient populations studied. Because patients receiving the same red marrow dose often experience different grades of toxicity, other factors might help predict the different grades of toxicity observed. One such factor may be the plasma FLT3-L (FMS-related tyrosine kinase 3 ligand, hematopoiesis stimulatory cytokine) level, which has been shown to be a better indicator of recovery of progenitor cells and, thus, red marrow radiosensitivity (because during the recovery period the progenitor cells are hyperproliferative and potentially more radiosensitive) for patients treated with previous chemotherapy than peripheral blood counts.
METHODS: Red marrow radiation doses were determined for 30 patients (20 male, 10 female; all without bone marrow or bone involvement; 19 had prior chemotherapy) after receiving (131)I RAIT (activity range, 2.1-8.9 GBq) for treatment of solid cancers known to produce carcinoembryonic antigen. Radiation dose estimates were calculated using 2 different methods of red marrow cumulated activity and red marrow-to-blood activity concentration ratio determinations for 2 dosimetric models: using both male and female and male-only masses and S values. Highest platelet toxicity grade at nadir (PTG), percentage platelet decrease (PPD) in counts, and platelet nadir (PN) counts were measured. FLT3-L levels (pg/mL) were determined by immunoassay before treatment; a normal FLT3-L level was assumed to be 80 pg/mL. The red marrow radiation doses (cGy) were adjusted for the patient's FLT3-L level when the patient's cytokine level exceeded the normal value. Marrow doses and FLT3-L-adjusted marrow doses were correlated with PTG, PPD, PN, and 1/PN. Administered activity, administered activity per unit body weight, and total body radiation dose were also correlated with these hematologic toxicity measures.
RESULTS: All red marrow dose calculation schemes resulted in essentially the same correlations for a given hematologic toxicity measure. Poor correlations were observed between administered activity, administered activity per unit body weight, total body radiation dose, or red marrow radiation dose and PTG, PPD, PN, and 1/PN. All correlations improved greatly when the various predictors of toxicity were adjusted for the patient's FLT3-L level. The highest correlation observed was between red marrow dose or total body dose and 1/PN (r = 0.86). Using an unadjusted red marrow dose to predict toxicity >/= grade 3, there were 8 true-positive, but 13 false-positive cases with 9 true-negatives. However, using a FLT3-L-adjusted red marrow dose, there were 8 true-positives, but only 2 false-positives and 20 true-negatives.
CONCLUSION: FLT3-L-adjusted red marrow radiation doses provide improved correlation with hematologic toxicity. Thus, elevated FLT3-L plasma levels before RAIT may indicate increased radiosensitivity of the bone marrow, and use of this measurement to adjust calculated red marrow or total body radiation doses may provide significantly better prediction of toxicity than radiation dose alone, leading to a patient-specific administered activity prescription that will deliver radiation doses to diseased tissues sufficient to produce an effective treatment outcome at acceptable toxicity levels.