This trial aims to improve hepatocellular carcinoma (HCC) tumor responses in patients undergoing Y90 radioembolization by using personalized dosimetry as part of treatment planning. Using standard calculations for Y90 doses may not be specific enough for individual patients given that there can be differences in how tumor cells and liver cells respond to radiation. Personalized dose plans may help improve treatment and outcomes in liver cancer.
PRIMARY OBJECTIVES: I. To achieve tumor objective response rates of 75% using voxel based dosimetry when the mean dose (Dmean) to tumor is targeted to be greater than 200 Gray (Gy). SECONDARY OBJECTIVES: I. Correlate changes in liver function and Common Terminology Criteria for Adverse Events (CTCAE) incidence to the mean absorbed dose to normal liver. II. Prospectively validate the accuracy of our published tumor dose response prediction based on the yttrium Y 90 glass microspheres (yttrium-90 \[Y90\]) tumor dose volume histograms (DVHs). III. Correlate predicted tumor doses macroaggregated albumin (MAA) scan with actual doses delivered (Y90 scan). IV. Develop a tumor dose response model prediction based on the MAA dose maps. V. Develop a model correlating normal liver radiation dose to liver function using single-photon emission computed tomography/computed tomography hepatobiliary iminodiacetic acid (SPECT/CT HIDA) imaging. VI. Develop a model correlating the relative tumor to normal liver enhancement on CT imaging and Cone beam CT imaging to the uptake on SPECT CT Tc99m MAA imaging (gold standard). VII. Compare tumor and normal liver doses estimations from SPECT CT Bremsstrahlung imaging to positron emission tomography (PET) CT imaging. OUTLINE: Patients undergo yttrium-90 microsphere radioembolization with yttrium Y 90 glass microspheres using personalized dose measurements. Patients also undergo SPECT/CT HIDA scan before radioembolization and 2-4 months after radioembolization. After completion of study treatment, patients are followed up at 3 and 6 months.
Study Type
INTERVENTIONAL
Allocation
NA
Purpose
TREATMENT
Masking
NONE
Enrollment
42
Undergo SPECT/CT HIDA scan
Undergo SPECT/CT HIDA scan
Undergo SPECT/CT HIDA scan
Undergo radioembolization with yttrium Y 90 glass microspheres
Undergo yttrium-90 microsphere radioembolization
M D Anderson Cancer Center
Houston, Texas, United States
Tumor objective response rate (ORR)
Assessed with modified Response Evaluation Criteria in Solid Tumors (mRECIST) criteria. The ORR of treated tumors will be determined. Will perform a cluster bootstrap in which patients (not tumors) are sampled with replacement. Will construct a 95% confidence interval using the 2.5% and 97.5% quantiles of the bootstrap sampling distribution.
Time frame: Up to 6 months
Changes in liver function
Will analyze the correlation in changes in liver function to the mean absorbed dose to normal liver. The hepatobiliary iminodiacetic acid (HIDA) scan will be used to estimate the body-surface area corrected mebrofenin clearance rate in %-activity/sec/m\^2 using published methodology. The change in liver function after radioembolization will be calculated and correlated to the radiation dose distribution to the normal liver. Will use Wilcoxon and two-sample t-tests to measure significance of association.
Time frame: Up to 6 months
Changes in Common Terminology Criteria for Adverse Events (CTCAE) incidence
Will analyze the correlation in changes in CTCAE incidence to the mean absorbed dose to normal liver. Will create waterfall charts of CTCAE and mean absorbed normal liver in liver. Will use Wilcoxon and two-sample t-tests to measure significance of association and scatterplots and regression (linear or non-linear) to predict actual doses delivered from predicted tumor dose.
Time frame: Up to 6 months
Accuracy of published tumor dose response prediction based on the yttrium-90 (Y90) tumor dose volume histograms
Lin's concordance correlation, t-test, and Bland-Altman analysis will be performed on pair-wise estimate of tumor doses. Develop tumor dose response curves using logistic regression and assess significance of parameters. Furthermore, we will compute the half-maximal dose (D50%), positive predictive value, negative predictive value and its 95% confidence interval from this prospective study. These values will be compared to those from the retrospective study for equivalence using the z-test.
Time frame: Up to 6 months
Prediction of tumor doses macroaggregated albumin (MAA) scan with actual doses delivered (Y90 scan)
Lin's concordance correlation, t-test, and Bland-Altman analysis will be performed on pair-wise estimate of tumor doses. Will use scatterplots and regression (linear or non-linear) to predict actual doses delivered from predicted tumor dose. A correlation analysis of the predicted tumor doses from the MAA images with the actual tumor doses delivered from Y90 images will be performed. The linear-correlation coefficient will be computed. A Bland-Altman analysis will be performed to compute the bias and the 95% limits of agreement.
Time frame: Up to 6 months
Development of tumor dose response model prediction based on MAA dose maps
Mean tumor doses based the planning 99mTc-MAA images will be computed using voxel dosimetry. A univariate logistic regression model will used to determine tumor dose metrics (D50%) that correlated with mRECIST-based tumor response at 3 and 6 months.
Time frame: Up to 6 months
Development of a model correlating normal liver radiation dose to liver function using single-photon emission computed tomography computed tomography hepatobiliary iminodiacetic acid (SPECT CT HIDA) imaging
A regression analysis of the change in liver function from the pre- and post-procedural HIDA SPECT/CT scans (changes in %/min) as a function of the normal liver absorbed dose will be performed.
Time frame: Up to 6 months
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.