Fifty-seventh annual meeting of the American association of physicists in medicine
TH-AB-BRA-12: Experimental Measurement of Microsphere Activity During Assay and Preparation Toward Establishing a Standard Methodology Amongst Vendors and Clinics
To measure real-time activity values for two manufacturers of Y-90 microspheres and experimentally determine the differences in activity values resulting from assay and preparation techniques; and furthermore to determine the influence microsphere solution homogeneity on these results.
A CRC-15BT dose calibrator was used to record activity values for Y-90 microspheres every 1000 ms. An I-125 ADCL calibrated seed was used to benchmark the assay technique. Measurements of Y-90 microspheres considered preparation with no-mixing to vigorously shaking for up to 60 seconds testing the effect of solution homogeneity. Activity values based on the microsphere settling time within the vial were contrasted with the activity corrected for decay during the measurement.
A review of 100+ pre-administration Y-90 microsphere delivery vials demonstrated differences between manufacturer supplied activities and in-house measurements of −3.1% to 2.0% with a standard deviation of 0.9% for vendor-A and −7.3% to +11.3% with a standard deviation of 3.3% for vendor-B. For vendor-A, the real-time assay of the recorded activity value was found to be independent of the amount of time the vial was shaken prior to measurement. For vendor-B, the recorded activity showed a dependence on the length of time that the delivery vial was shaken prior to assay resulting in an equilibrium time of approximately 144 seconds for mixing times ≥ 15 seconds. However, this effect is small (∼0.5%) and be considered technician/operator independent.
Activity values provided by the vendor and those measured at the local clinic can be as large as ±10% and consequently could have significant downstream dosimetric effects. Uncertainties could be reduced if a NIST/ADCL traceable calibration was provided by the vendor. Vendor- A provides a certificate while Vendor-B does not resulting in significantly larger uncertainties. These uncertainties are shown to be independent of the assay technique and entirely vendor dependent.