Development and Validation of a Small Animal Immobilizer and Positioning System for the Study of Delivery of Intracranial and Extracranial Radiotherapy Using the Gamma Knife System.

The purpose of this research is to establish a process of irradiating mice using the Gamma Knife as a versatile system for small animal irradiation and to validate accurate intracranial and extracranial dose delivery using this system. A stereotactic immobilization device was developed for small animals for the Gamma Knife head frame allowing for isocentric dose delivery. Intercranial positional reproducibility of a reference point from a primary reference animal was verified on an additional mouse. Extracranial positional reproducibility of the mouse aorta was verified using 3 mice. Accurate dose delivery was validated using film and thermoluminescent dosimeter measurements with a solid water phantom. Gamma Knife plans were developed to irradiate intracranial and extracranial targets. Mice were irradiated validating successful targeted radiation dose delivery. Intramouse positional variability of the right mandible reference point across 10 micro-computed tomography scans was 0.65 ± 0.48 mm. Intermouse positional reproducibility across 2 mice at the same reference point was 0.76 ± 0.46 mm. The accuracy of dose delivery was 0.67 ± 0.29 mm and 1.01 ± 0.43 mm in the coronal and sagittal planes, respectively. The planned dose delivered to a mouse phantom was 2 Gy at the 50% isodose with a measured thermoluminescent dosimeter dose of 2.9 ± 0.3 Gy. The phosphorylated form of member X of histone family H2A (γH2AX) staining of irradiated mouse brain and mouse aorta demonstrated adjacent tissue sparing. In conclusion, our system for preclinical studies of small animal irradiation using the Gamma Knife is able to accurately deliver intracranial and extracranial targeted focal radiation allowing for preclinical experiments studying focal radiation.

Development of a Monte Carlo model for treatment planning dose verification of the Leksell Gamma Knife Perfexion radiosurgery system.

Detailed Monte Carlo (MC) modeling of the Leksell Gamma Knife (GK) Perfexion (PFX) collimator system is the only accurate ab initio approach appearing in the literature. As a different approach, in this work, we present a MC model based on film measurement. By adjusting the model parameters and fine-tuning the derived fluence map for each individual source to match the manufacturer's ring output factors, we created a reasonable virtual source model for MC simulations to verify treatment planning dose for the GK PFX radiosurgery system. The MC simulation model was commissioned by simple single shots. Dose profiles and both ring and collimator output factors were compared with the treatment planning system (TPS). Good agreement was achieved for dose profiles especially for the region of plateau (< 2%), while larger difference (< 5%) came from the penumbra region. The maximum difference of the calculated output factor was within 0.7%. The model was further validated by a clinical test case. Good agreement was obtained. The DVHs for brainstem and the skull were almost identical and, for the target, the volume covered by the prescription (12.5 Gy to 50% isodose line) was 95.6% from MC calculation versus 100% from the TPS.

Experimental Validation of Monte Carlo Simulations Based on a Virtual Source Model for TomoTherapy in a RANDO Phantom.

A virtual source model for Monte Carlo simulations of helical TomoTherapy has been developed previously by the authors. The purpose of this work is to perform experiments in an anthropomorphic (RANDO) phantom with the same order of complexity as in clinical treatments to validate the virtual source model to be used for quality assurance secondary check on TomoTherapy patient planning dose. Helical TomoTherapy involves complex delivery pattern with irregular beam apertures and couch movement during irradiation. Monte Carlo simulation, as the most accurate dose algorithm, is desirable in radiation dosimetry. Current Monte Carlo simulations for helical TomoTherapy adopt the full Monte Carlo model, which includes detailed modeling of individual machine component, and thus, large phase space files are required at different scoring planes. As an alternative approach, we developed a virtual source model without using the large phase space files for the patient dose calculations previously. In this work, we apply the simulation system to recompute the patient doses, which were generated by the treatment planning system in an anthropomorphic phantom to mimic the real patient treatments. We performed thermoluminescence dosimeter point dose and film measurements to compare with Monte Carlo results. Thermoluminescence dosimeter measurements show that the relative difference in both Monte Carlo and treatment planning system is within 3%, with the largest difference less than 5% for both the test plans. The film measurements demonstrated 85.7% and 98.4% passing rate using the 3 mm/3% acceptance criterion for the head and neck and lung cases, respectively. Over 95% passing rate is achieved if 4 mm/4% criterion is applied. For the dose-volume histograms, very good agreement is obtained between the Monte Carlo and treatment planning system method for both cases. The experimental results demonstrate that the virtual source model Monte Carlo system can be a viable option for the accurate dose calculation of helical TomoTherapy.

Investigation of Nonuniform Dose Voxel Geometry in Monte Carlo Calculations.

The purpose of this work is to investigate the efficacy of using multi-resolution nonuniform dose voxel geometry in Monte Carlo (MC) simulations. An in-house MC code based on the dose planning method MC code was developed in C++ to accommodate the nonuniform dose voxel geometry package since general purpose MC codes use their own coupled geometry packages. We devised the package in a manner that the entire calculation volume was first divided into a coarse mesh and then the coarse mesh was subdivided into nonuniform voxels with variable voxel sizes based on density difference. We name this approach as multi-resolution subdivision (MRS). It generates larger voxels in small density gradient regions and smaller voxels in large density gradient regions. To take into account the large dose gradients due to the beam penumbra, the nonuniform voxels can be further split using ray tracing starting from the beam edges. The accuracy of the implementation of the algorithm was verified by comparing with the data published by Rogers and Mohan. The discrepancy was found to be 1% to 2%, with a maximum of 3% at the interfaces. Two clinical cases were used to investigate the efficacy of nonuniform voxel geometry in the MC code. Applying our MRS approach, we started with the initial voxel size of 5 × 5 × 3 mm(3), which was further divided into smaller voxels. The smallest voxel size was 1.25 × 1.25 × 3 mm(3). We found that the simulation time per history for the nonuniform voxels is about 30% to 40% faster than the uniform fine voxels (1.25 × 1.25 × 3 mm(3)) while maintaining similar accuracy.

Establishing a process of irradiating small animal brain using a CyberKnife and a microCT scanner.

PURPOSE: Establish and validate a process of accurately irradiating small animals using the CyberKnife G4 System (version 8.5) with treatment plans designed to irradiate a hemisphere of a mouse brain based on microCT scanner images. METHODS: These experiments consisted of four parts: (1) building a mouse phantom for intensity modulated radiotherapy (IMRT) quality assurance (QA), (2) proving usability of a microCT for treatment planning, (3) fabricating a small animal positioning system for use with the CyberKnife's image guided radiotherapy (IGRT) system, and (4)in vivo verification of targeting accuracy. A set of solid water mouse phantoms was designed and fabricated, with radiochromic films (RCF) positioned in selected planes to measure delivered doses. After down-sampling for treatment planning compatibility, a CT image set of a phantom was imported into the CyberKnife treatment planning system--MultiPlan (ver. 3.5.2). A 0.5 cm diameter sphere was contoured within the phantom to represent a hemispherical section of a mouse brain. A nude mouse was scanned in an alpha cradle using a microCT scanner (cone-beam, 157 × 149 pixels slices, 0.2 mm longitudinal slice thickness). Based on the results of our positional accuracy study, a planning treatment volume (PTV) was created. A stereotactic body mold of the mouse was "printed" using a 3D printer laying UV curable acrylic plastic. Printer instructions were based on exported contours of the mouse's skin. Positional reproducibility in the mold was checked by measuring ten CT scans. To verify accurate dose delivery in vivo, six mice were irradiated in the mold with a 4 mm target contour and a 2 mm PTV margin to 3 Gy and sacrificed within 20 min to avoid DNA repair. The brain was sliced and stained for analysis. RESULTS: For the IMRT QA using a set of phantoms, the planned dose (6 Gy to the calculation point) was compared to the delivered dose measured via film and analyzed using Gamma analysis (3% and 3 mm). A passing rate of 99% was measured in areas of above 40% of the prescription dose. The final inverse treatment plan was comprised of 43 beams ranging from 5 to 12.5 mm in diameter (2.5 mm size increments are available up to 15 mm in diameter collimation). Using the Xsight Spine Tracking module, the CyberKnife system could not reliably identify and track the tiny mouse spine; however, the CyberKnife system could identify and track the fiducial markers on the 3D mold.In vivo positional accuracy analysis using the 3D mold generated a mean error of 1.41 mm ± 0.73 mm when fiducial markers were used for position tracking. Analysis of the dissected brain confirmed the ability to target the correct brain volume. CONCLUSIONS: With the use of a stereotactic body mold with fiducial markers, microCT imaging, and resolution down-sampling, the CyberKnife system can successfully perform small-animal radiotherapy studies.

MIRD pamphlet No. 24: Guidelines for quantitative 131I SPECT in dosimetry applications.

The reliability of radiation dose estimates in internal radionuclide therapy is directly related to the accuracy of activity estimates obtained at each imaging time point. The recently published MIRD pamphlet no. 23 provided a general overview of quantitative SPECT imaging for dosimetry. The present document is the first in a series of isotope-specific guidelines that will follow MIRD 23 and focuses on one of the most commonly used therapeutic radionuclides, (131)I. The purpose of this document is to provide guidance on the development of protocols for quantitative (131)I SPECT in radionuclide therapy applications that require regional (normal organs, lesions) and 3-dimensional dosimetry.

Phase II trial of radiosurgery to magnetic resonance spectroscopy-defined high-risk tumor volumes in patients with glioblastoma multiforme.

PURPOSE: To determine the efficacy of a Gamma Knife stereotactic radiosurgery (SRS) boost to areas of high risk determined by magnetic resonance spectroscopy (MRS) functional imaging in addition to standard radiotherapy for patients with glioblastoma (GBM). METHODS AND MATERIALS: Thirty-five patients in this prospective Phase II trial underwent surgical resection or biopsy for a GBM followed by SRS directed toward areas of MRS-determined high biological activity within 2 cm of the postoperative enhancing surgical bed. The MRS regions were determined by identifying those voxels within the postoperative T2 magnetic resonance imaging volume that contained an elevated choline/N-acetylaspartate ratio in excess of 2:1. These voxels were marked, digitally fused with the SRS planning magnetic resonance image, targeted with an 8-mm isocenter per voxel, and treated using Radiation Therapy Oncology Group SRS dose guidelines. All patients then received conformal radiotherapy to a total dose of 60 Gy in 2-Gy daily fractions. The primary endpoint was overall survival. RESULTS: The median survival for the entire cohort was 15.8 months. With 75% of recursive partitioning analysis (RPA) Class 3 patients still alive 18 months after treatment, the median survival for RPA Class 3 has not yet been reached. The median survivals for RPA Class 4, 5, and 6 patients were 18.7, 12.5, and 3.9 months, respectively, compared with Radiation Therapy Oncology Group radiotherapy-alone historical control survivals of 11.1, 8.9, and 4.6 months. For the 16 of 35 patients who received concurrent temozolomide in addition to protocol radiotherapeutic treatment, the median survival was 20.8 months, compared with European Organization for Research and Treatment of Cancer historical controls of 14.6 months using radiotherapy and temozolomide. Grade 3/4 toxicities possibly attributable to treatment were 11%. CONCLUSIONS: This represents the first prospective trial using selective MRS-targeted functional SRS combined with radiotherapy for patients with GBM. This treatment is feasible, with acceptable toxicity and patient survivals higher than in historical controls. This study can form the basis for a multicenter, randomized trial.

An automatic method for PET target segmentation using a lookup table based on volume and concentration ratio.

Accurate evaluation of functionally significant target volumes in combination with anatomic imaging is of primary importance for effective radiation therapy treatment planning. In this study, a method for rapid and accurate PET image segmentation and volumetrics based on phantom measurements and independent of scanner calibration was developed. A series of spheres ranging in volume from 0.5 mL to 95 mL were imaged in an anthropomorphic phantom of human thorax using two commercial PET and CT/PET scanners. The target to background radioactivity concentration ratio ranged from 3:1 to 12:1 in 11 separate phantom scanning experiments. The results confirmed that optimal segmentation thresholding depends on target volume and radioactivity concentration ratio. This information can be derived from a generalized pre-determined "lookup table" of volume and contrast dependent threshold values instead of using fitted curves derived from machine specific information. A three-step method based on the PET image intensity information alone was used to delineate volumes of interest. First, a mean intensity segmentation method was used to generate an initial estimate of target volume, and the radioactivity concentration ratio was computed by a family of recovery coefficient curves to compensate for the partial volume effect. Next, the appropriate threshold value was obtained from a phantom-generated threshold lookup table. Lastly, a threshold level set method was performed on the threshold value to further refine the target contour by reducing the limitation of global thresholding. The segmentation results were consistent for spheres greater than 2.5 mL which yielded volume average uncertainty of 11.2% in phantom studies. The results of segmented volumes were comparable to those determined by contrast-oriented method and iterative threshold method (ITM). In addition, the new volume segmentation method was applied clinically to ten patients undergoing PET/CT volume analysis for radiation therapy treatment planning of solitary lung metastases. For these patients, the average PET segmented volumes were within 8.0% of the CT volumes and were highly dependent on the extension of functionally inactive tumor volume. In summary, the current method does not require fitted threshold curves or a priori knowledge of the CT/MRI target volume. This threshold method can be universally applied to radiation therapy treatment planning with comparable accuracy, and may be useful in the rapid identification and assessment of plans containing multiple targets.

MIRD Pamphlet No. 22 (abridged): radiobiology and dosimetry of alpha-particle emitters for targeted radionuclide therapy.

The potential of alpha-particle emitters to treat cancer has been recognized since the early 1900s. Advances in the targeted delivery of radionuclides and radionuclide conjugation chemistry, and the increased availability of alpha-emitters appropriate for clinical use, have recently led to patient trials of radiopharmaceuticals labeled with alpha-particle emitters. Although alpha-emitters have been studied for many decades, their current use in humans for targeted therapy is an important milestone. The objective of this work is to review those aspects of the field that are pertinent to targeted alpha-particle emitter therapy and to provide guidance and recommendations for human alpha-particle emitter dosimetry.

MIRD pamphlet No. 20: the effect of model assumptions on kidney dosimetry and response--implications for radionuclide therapy.

UNLABELLED: Renal toxicity associated with small-molecule radionuclide therapy has been shown to be dose-limiting for many clinical studies. Strategies for maximizing dose to the target tissues while sparing normal critical organs based on absorbed dose and biologic response parameters are commonly used in external-beam therapy. However, radiopharmaceuticals passing though the kidneys result in a differential dose rate to suborgan elements, presenting a significant challenge in assessing an accurate dose-response relationship that is predictive of toxicity in future patients. We have modeled the multiregional internal dosimetry of the kidneys combined with the biologic response parameters based on experience with brachytherapy and external-beam radiation therapy to provide an approach for predicting radiation toxicity to the kidneys. METHODS: The multiregion kidney dosimetry model of MIRD pamphlet no. 19 has been used to calculate absorbed dose to regional structures based on preclinical and clinical data. Using the linear quadratic model for radiobiologic response, we computed regionally based surviving fractions for the kidney cortex and medulla in terms of their concentration ratios for several examples of radiopharmaceutical uptake and clearance. We used past experience to illustrate the relationship between absorbed dose and calculated biologically effective dose (BED) with radionuclide-induced nephrotoxicity. RESULTS: Parametric analysis for the examples showed that high dose rates associated with regions of high activity concentration resulted in the greatest decrease in tissue survival. Higher dose rates from short-lived radionuclides or increased localization of radiopharmaceuticals in radiosensitive kidney subregions can potentially lead to greater whole-organ toxicity. This finding is consistent with reports of kidney toxicity associated with early peptide receptor radionuclide therapy and (166)Ho-phosphonate clinical investigations. CONCLUSION: Radionuclide therapy dose-response data, when expressed in terms of biologically effective dose, have been found to be consistent with external-beam experience for predicting kidney toxicity. Model predictions using both the multiregion kidney and linear quadratic models may serve to guide the investigator in planning and optimizing future clinical trials of radionuclide therapy.

Cyberknife radiosurgery for squamous cell carcinoma of vulva after prior pelvic radiation therapy.

Limited options exist for patients experiencing a local recurrence of vulvar malignancies after surgery and pelvic radiation. These recurrences often are associated with cancer-related skin desquamation and poor clinical outcomes. A new radiotherapeutic treatment modality for the previously irradiated patient is cyberknife radiosurgery, which uses a linear accelerator mounted on an industrial robotic arm to allow non-coplanar radiation therapy delivery with sub-millimeter precision. This study describes the first reported use of cyberknife radiosurgery for the treatment of recurrent vulvar cancer in three women.

Risks to normal tissues from radionuclide therapy.

The development of radionuclide therapies during the past few decades provides a growing body of data on radiobiologic effects, including normal tissue toxicities and antitumor efficacy. Information on normal tissue toxicity from radionuclides is more limited than that from external beam radiation and appears to be more variable. Much of the increased variability is attributed to heterogeneous distribution, which complicates the potential for whole-organ toxicity, and the differences in dosimetry methodology. Although new tools are becoming available, quantitation of heterogeneous dose for radionuclides is usually less precise than dosimetry that is used in external beam radiation practice. The correlation between reported dose estimates and toxicity has improved during the past 2 decades, partly as the result of increased accuracy and standardization of dosimetry techniques and to adjustment for biologic effects. This review provides an updated compendium of dose-response relationships and consideration of dosimetry as well as radiobiologic factors that influence the reported results. Data presented are mainly derived from studies involving deliver of radiation to adults with malignancies, with most experience from radionuclides that predominantly emit beta radiation.

Overview of dosimetry for Systemic Targeted Radionuclide Therapy (STaRT).

The purposes of systemic targeted radionuclide therapy dosimetry include compiling a database of normal organ radiation-absorbed doses that are carrier- and radionuclide-specific, and assuring that the normal organ radiation doses are within a safe range before therapy. Also of importance is quantitation of radiation delivery to tumors vs. normal tissues to correlate absorbed dose with tumor control. For agents with significant and variable excretion, estimates of individual patient distribution/clearance may be needed to optimize the dose-response relationship.

Phase I study of 131I-chimeric(ch) TNT-1/B monoclonal antibody for the treatment of advanced colon cancer.

PURPOSE: The primary aim of this study was to evaluate the biodistribution and toxicity of 131I-chimeric(ch) TNT-1/B monoclonal antibody (MAB), which binds to intracellular antigens of necrotic regions within tumors, in patients with advanced colon or colorectal cancer. The rationale for targeting areas of tumor necrosis is the observation that necrotic lesions are more abundant in cancer lesions than in surrounding tissues. PATIENTS AND METHODS: Cohorts of patients with advanced colon or colorectal cancer were administered a one-time 30-60-minute intravenous (i.v.) infusion of 131I-chTNT-1/B at doses ranging from 12.95 to 66.23 MBq/kg (0.35-1.79 mCi/kg). RESULTS: The dose-limiting toxicity, experienced at 66.23 MBq/kg (1.79 mCi/kg) 131I-chTNT-1/B MAB, was myelosuppression. Two (2) patients at the 66.23-MBq/kg (1.79 mCi/kg) dose level had both grade 3 thrombocytopenia and grade 3 neutropenia that persisted for at least 2 weeks but were reversible. The maximum tolerated dose was 58.09 MBq/kg (1.57 mCi/kg) 131I-chTNT-1/B MAB. Of the 21 patients, one developed a moderate human antichimeric antibody (HACA) response and 6 developed low HACA responses. CONCLUSIONS: The infusion of 131I-chTNT-1/B MAB was well tolerated, without significant nonhematological toxicity. No patient obtained a complete or partial response, based on tumor cross-product response criteria. Tumor localization was seen in patients with dose levels at, and exceeding, 50.23 MBq/kg (1.36 mCi/kg) 131I-chTNT-1/B MAB.

12 Gy gamma knife radiosurgical volume is a predictor for radiation necrosis in non-AVM intracranial tumors.

PURPOSE: To determine whether the 12-Gy radiosurgical volume (12-GyV) correlates with the development of postradiosurgical imaging changes suggestive of radiation necrosis in patients treated for non-arteriovenous malformation (non-AVM) intracranial tumors with gamma knife stereotactic radiosurgery (GKSRS). METHODS AND MATERIALS: A retrospective single-institution review of 129 patients with 198 separate non-AVM tumors was performed. Patients were followed with magnetic resonance imaging (MRI) and physical examinations at 3- to 6-month intervals. Patients who developed postradiosurgical MRI changes suggestive of radiation necrosis were labeled as having either symptomatic radiation necrosis (S-NEC) if they experienced any decline in neurologic examination associated with the imaging changes, or asymptomatic radiation necrosis (A-NEC) if they had a stable or improving neurologic examination. RESULTS: 12-GyV correlated with risk of S-NEC, which was 23% (for 12-GyV of 0-5 cc), 20% (5-10 cc), 54% (10-15 cc), and 57% (>15 cc). The risk of A-NEC did not significantly change with 12-GyV. Logistic regression analyses showed that the following factors were associated with the development of S-NEC: 12-GyV (p<0.01), occipital and temporal lesions (p<0.01), previous whole-brain radiotherapy (p=0.03), and male sex (p=0.03). Radiosurgical plan conformality did not correlate with the development of S-NEC. CONCLUSION: The risk of S-NEC, but not A-NEC after GKSRS for non-AVM tumors correlates with 12-GyV, and increases significantly for 12-GyV>0 cc.

Safety and feasibility of convection-enhanced delivery of Cotara for the treatment of malignant glioma: initial experience in 51 patients.

OBJECTIVE: We report the safety and feasibility of using convection-enhanced delivery to administer Cotara (Peregrine Pharmaceuticals, Inc., Tustin, CA), a novel radioimmunotherapeutic agent, to patients with malignant glioma. METHODS: Between April 1998 and November 2002, 51 patients with histologically confirmed malignant glioma received Cotara by convection-enhanced delivery. Most patients (88%) were treated with Cotara targeting tumor volume-dependent, single or multiple administrations of activity ranging from 0.5 to 3.0 mCi/cm3 of baseline clinical target volume. Two weeks after infusion, single-photon emission computed tomographic imaging determined the spatial distribution of Cotara. Patients were followed for as long as 41 months (average follow-up, 5 mo). Safety was evaluated on the basis of incidence of procedure-related, neurological, and systemic adverse events. Feasibility was evaluated in a subset of patients on the basis of the correlation between the prescribed activity and the actual activity administered to the targeted region. RESULTS: Fifty-one patients, 37 with recurrent glioblastoma multiforme, 8 with newly diagnosed glioblastoma multiforme, and 6 with recurrent anaplastic astrocytomas, were treated. Average tumor volume was 36 +/- 27.6 cm3 (range, 5-168 cm3). Of the 67 infusions, 13 (19%), 52 (78%), and 2 (3%) delivered less than 90%, 100 +/- 10%, and more than 110%, respectively, of the prescribed administered activity to the targeted region. Treatment-emergent, drug-related central nervous system adverse events included brain edema (16%), hemiparesis (14%), and headache (14%). Systemic adverse events were mild. Several patients had objective responses to Cotara. CONCLUSION: The majority of Cotara infusions delivered between 90 and 110% of the prescribed administered activity to the targeted region. This method of administration has an acceptable safety profile compared with literature reports of other therapeutics delivered by convection-enhanced delivery.

Bone marrow dosimetry using blood-based models for radiolabeled antibody therapy: a multiinstitutional comparison.

UNLABELLED: Standardization of marrow dosimetry is of considerable importance when estimating dose-response for a multicentered clinical trial involving radionuclide therapy. However, it is only within the past five years that the intercomparison of marrow dosimetry results among separate clinical trials that use the same agent has become scientifically feasible. In this work, we have analyzed reported marrow dosimetry results from radioimmunotherapy trials and recalculated marrow absorbed doses at a central facility using a standard blood model with patient-specific source data. The basic approach used in the American Association of Physicists in Medicine (AAPM)/Sgouros marrow dosimetry methodology was common to calculation performed at all participating institutions, including the central facility. Differences in dose estimates associated with starting assumptions and the exact implementation of the AAPM/Sgouros calculation methodology used by the source institutions and the central facility were quantified and compared. METHODS: Data from 22 patients enrolled in radiolabeled antibody clinical trials were randomly selected from 7 participating institutions for the assessment of marrow dose. The analysis was restricted to those patients who were treated with 131I- or 186Re-labeled antibody and had no marrow involvement. Calculation of bone marrow dose at each participating institution was unique to the trial or institution, but all used some form of the AAPM/Sgouros blood model approach. The central facility adopted a marrow dosimetry model based on the AAPM/Sgouros model for radiolabeled antibodies using the standard MIRD approach to the remainder-of-body contribution. A standardized approach to account for variations in patient mass was used for the remainder-of-body component. To simplify clinical implementation, regional marrow uptake and time-dependent changes in the marrow-to-blood concentration ratio were not included. Methods of formatting the collection of standard datasets useful in defining dose-response parameters are also presented. RESULTS: Bone marrow doses were calculated according to the method described for each of the 22 patients based on the patient-specific data supplied by the participating institutions. These values were then individually compared with the marrow doses originally reported by each institution. Comparison of the two calculation methods was expressed as a ratio of the marrow doses for each patient. The mean ratio for the dose estimates at the participating institution calculation compared with the central laboratory value was 0.920 +/- 0.259 (mean +/- SD), with a range from 0.708 to 1.202. CONCLUSION: The independent use of the AAPM/Sgouros method blood model approach to marrow dosimetry has brought these dose estimates to within 30% of the results obtained centrally compared with substantially higher uncertainties reported previously. Variations in calculation methodology or initial assumptions adopted by individual institutions may still contribute significant uncertainty to dose estimates, even when the same data are used as a starting point for the calculation comparison shown here. A clinically relevant, standard method for marrow dosimetry for radiolabeled antibodies is proposed as a benchmark for intercomparison purposes. A parameter sensitivity analysis and a summary discussion of the use of this model for potentially improving dose-response data correlation are also presented.