Single-Cell Dosimetry for Radioimmunotherapy of B-Cell Lymphoma Patients with Special Reference to Leukemic Spread

Cellular S-value evaluation based on real human cell models using the GATE MC package

Exploring the spatial distribution of the energy loss of ionising radiation at the subcellular level is indispensable for evaluating the radiobiological effects of targeted radionuclide therapy accurately. Believing that S-values are important for obtaining the target dose, the Committee on Medical Internal Radiation Dose (MIRD) proposed a method to obtain the cellular dosimetric parameter. However, most available data on cellular S-values were calculated based on simple geometric models, such as ellipsoids or spheres, which do not accurately reflect biological reality. To investigate the influence of the cellular model on S-values, calculations were performed for two kinds of polygon-surface phantom models of realistic, individual human cells, the lung epithelial cell model (the B2B Phantom model) and the hepatocyte model (the Liver Phantom model), using the Monte Carlo (MC) software package GATE. To analyse the influence of cell geometry on the final S-value, the differences in the S-values between the realistic cell models and simple geometric sphere and ellipsoid models with similar volumes were calculated and compared for six different combinations of source and target regions. The irradiation conditions were 0.01-1.10 MeV monoenergetic electron sources and the Auger electronic therapy nuclides Ga-67, Tc-99m, In-111, I-125 and Tl-201, which are commonly used in nuclear medicine. The S-values calculated in this study are different from the results of the simple geometry models proposed by previous researchers. Two more precise polygon-surface phantom models of realistic, individual human cells were used, which provided more accurate information about the cell dose and will be very useful for the diagnostic application of radiotherapy in the future.

Monte Carlo simulation of electron dose from (131)I-targeted tumor cells within a heterogeneous tumor

BACKGROUND AND AIM

In internal radiotherapy, the variable distribution of target receptors within the tumoral tissue, and the variable ranges of electrons may be responsible for a heterogeneous dose distribution at the cellular level. The aim of the present study was to use Monte Carlo simulations to assess (131)I electron dose in a model of heterogeneous tumor containing multiple clusters of cancer cells, targeted by (131)I-labeled molecules.

METHODS

The model consisted of 150-μm-diameter spherical tumor cell clusters, in which (131)I was homogeneously distributed. Clusters were placed 24 μm apart, separated by septa of nonradioactive connective tissue. The electron dose distribution to tumor cells in a single cluster was first assessed. Then was assessed the dose increase to these targets after adding multiple layers of neighboring clusters (total number of clusters = 15,624).

RESULTS

Dose distribution within a single isolated cluster follows a decreasing gradient, the dose for the outermost cell layer being about half that at the center. When radioactive neighbors were added, the dose to the central cluster increased. The most important contribution was given by the nearest neighbors, whereas the contribution from neighbors beyond a distance of 1 mm was only for 5% of the final dose. If the central cluster was unlabeled, the absorbed dose to the outermost cell layer of this cluster was reduced by 27%, and that at the center by 45%.

CONCLUSIONS

The electron cross-dose of (131)I falls rapidly as a function of distance and becomes negligible after just 1 mm. Small clusters of tumor cells that are not radiolabeled may receive a very small dose. Therefore, in internal radiotherapy it is important to aim at targeting tumor cells as homogeneously as possible, rather than relying on the cross-dose to achieve a therapeutic effect.

Dosimetry on sub-cellular level for intracellular incorporated auger-electron-emitting radionuclides: a comparison of Monte Carlo simulations and analytic calculations

A quantitative dosimetric comparison was performed between Monte Carlo (MC) simulations and analytic calculations at the (sub) cellular level (V79 cells) for four nucleus-incorporated radiochemicals ((125)I/(123)I/(77)Br-UdR and A (125)IP) and two radiochemicals that localised mainly in the cytoplasm of cells ((125)I-dihydrorhodamine and Na(2)(51)CrO(4)). A microscopic investigation around the decay site of the three DNA-incorporated radionuclides ((125)I/(123)I/(77)Br-UdR) was also carried out. On the whole, deviations between MC and analytic calculations for the absorbed dose and dose rate to the cell nucleus were within ∼10%. The dose rate to the nucleus for the radiochemicals that mainly localised in the cytoplasm was greater than that for the nucleus-incorporated ones. Also evident was that the dose rate to the nucleus was approximately the same for the three DNA-incorporated radiochemicals. In contrast to the small differences found between MC and analytic calculations for the (average) absorbed dose to the nucleus, the dosimetric analysis at the microscopic level for the three DNA-incorporated radionuclides showed that the two computational approaches lead to a completely different energy deposition pattern around the decay site.

Hexa-arginine enhanced uptake and residualization of selective high affinity ligands by Raji lymphoma cells

Background

A variety of arginine-rich peptide sequences similar to those found in viral proteins have been conjugated to other molecules to facilitate their transport into the cytoplasm and nucleus of targeted cells. The selective high affinity ligand (SHAL) (DvLPBaPPP)₂LLDo, which was developed to bind only to cells expressing HLA-DR10, has been conjugated to one of these peptide transduction domains, hexa-arginine, to assess the impact of the peptide on SHAL uptake and internalization by Raji cells, a B-cell lymphoma.

Results

An analog of the SHAL (DvLPBaPPP)₂LLDo containing a hexa-arginine peptide was created by adding six D-arginine residues sequentially to a lysine inserted in the SHAL's linker. SHAL binding, internalization and residualization by Raji cells expressing HLA-DR10 were examined using whole cell binding assays and confocal microscopy. Raji cells were observed to bind two fold more ¹¹¹In-labeled hexa-arginine SHAL analog than Raji cells treated with the parent SHAL. Three fold more hexa-arginine SHAL remained associated with the Raji cells after washing, suggesting that the peptide also enhanced residualization of the ¹¹¹In transported into cells. Confocal microscopy showed both SHALs localized in the cytoplasm of Raji cells, whereas a fraction of the hexa-arginine SHAL localized in the nucleus.

Conclusion

The incorporation of a hexa-D-arginine peptide into the linker of the SHAL (DvLPBaPPP)₂LLDo enhanced both the uptake and residualization of the SHAL analog by Raji cells. In contrast to the abundant cell surface binding observed with Lym-1 antibody, the majority of (DvLPBaPPP)₂LArg6AcLLDo and the parent SHAL were internalized. Some of the internalized hexa-arginine SHAL analog was also associated with the nucleus. These results demonstrate that several important SHAL properties, including uptake, internalization, retention and possibly intracellular distribution, can be enhanced or modified by conjugating the SHALs to a short polypeptide.

Calculation of electron dose to target cells in a complex environment by Monte Carlo code "CELLDOSE"

BACKGROUND

We used the Monte Carlo code "CELLDOSE" to assess the dose received by specific target cells from electron emissions in a complex environment. (131)I in a simulated thyroid was used as a model.

METHODS

Thyroid follicles were represented by 170 microm diameter spherical units made of a lumen of 150 microm diameter containing colloidal matter and a peripheral layer of 10 microm thick thyroid cells. Neighbouring follicles are 4 microm apart. (131)I was assumed to be homogeneously distributed in the lumen and absent in cells. We firstly assessed electron dose distribution in a single follicle. Then, we expanded the simulation by progressively adding neighbouring layers of follicles, so to reassess the electron dose to this single follicle implemented with the contribution of the added layers.

RESULTS

Electron dose gradient around a point source showed that the (131)I electron dose is close to zero after 2,100 microm. Therefore, we studied all contributions to the central follicle deriving from follicles within 12 orders of neighbourhood (15,624 follicles surrounding the central follicle). The dose to colloid of the single follicle was twice as high as the dose to thyroid cells. Even when all neighbours were taken into account, the dose in the central follicle remained heterogeneous. For a 1-Gy average dose to tissue, the dose to colloidal matter was 1.168 Gy, the dose to thyroid cells was 0.982 Gy, and the dose to the inter-follicular tissue was 0.895 Gy. Analysis of the different contributions to thyroid cell dose showed that 17.3% of the dose derived from the colloidal matter of their own follicle, while the remaining 82.7% was delivered by the surrounding follicles. On the basis of these data, it is shown that when different follicles contain different concentrations of (131)I, the impact in terms of cell dose heterogeneity can be important.

CONCLUSION

By means of (131)I in the thyroid as a theoretical model, we showed how a Monte Carlo code can be used to map electron dose deposit and build up the dose to target cells in a complex multi-source environment. This approach can be of considerable interest for comparing different radiopharmaceuticals as therapy agents in oncology.