Difficulties with reproducibility of isolation yields when distilling 211At from irradiated bismuth targets led us to use a “wet chemistry” approach for that process1. The wet chemistry approach has provided 211At isolation yields of ~ 78 % after decay and Bi attenuation corrections2. However, the use of diisopropyl ether (DIPE) in the separation process has made it difficult to reach our goal of automating the 211At isolation. Therefore, we have investigated the use of column materials to simplify the isolation of 211At and remove DIPE from the process. In this investigation we evaluated the use of a strong anion exchange resin (AG1×8), a strong cation exchange resin (AG MP-50) and a polyethylene glycol (PEG)-coated resin for separation of 211At from the bismuth target material. Anion and cation resins AG1×8 and AG MP-50 were obtained from commercial sources. A PEG-coated resin was prepared by reaction of the Merrifield resin with mPEG-OH 2000 in the pres-ence of tBuOK at 80 °C for 3 days, followed by drying under vacuum. Prior to use of the PEG resin, it was soaked in H2O. Resins (400–800 mg) were loaded into polypropylene columns (Applied Separations, Inc.). Column elution studies were conducted with and without reductants (0.75M FeSO4/1M H2SO4 or Na2S2O5) to determine their effect on capture of 211At. After target dissolution in HNO3 (and in most cases subse-quent removal of HNO3 by distillation and redis-solution of solid in 8M HCl), 211At solution was loaded onto the column, then the column was washed with 2M HCl or H2O to separate the Bi, and finally was eluted with strong base to remove the 211At. Initial studies were conducted with stable iodine to determine if reductants were effective in the presence of large amounts of bismuth ions. Studies with AG1×8 used 125I to determine if that radiohalogen could be captured and recovered from the column when eluting with boric acid buffers at pH 5.3, 8.0 or 10, or H2O at pH 7. Capture and recovery of 211At was evaluated under the same conditions. Further studies with AG1×8 involved eluting with 4M H2SO4. A limited study with AG MP-50 resin used 1M HCl as eluant. Studies with PEG-coated columns used 2M HCl, 4M HCl, 8M HCl, 16 M HNO3 and 8M HNO3 as initial (capture) eluants. Strong base (0.2, 1 or 12.5 M NaOH; 15M NH4OH) and 3 or 500 mM tetrabutylammonium bromide (TBAB) were evaluated for removal of 211At from the columns tested. The efficiency for capture of 211At on the AG1×8 column was high (99%) when loading with strong acid, but decreased when using 0.1–0.2M boric acid (69–91 %) buffer. Low 211At capture efficiencies were obtained with AG MP-50 col-umns (15–29%). High 211At capture efficiencies (96–100%) were obtained with PEG-coated resins when loading with 8M HCl or 8M HNO3, irre-spective of whether reductant was in the acid solution. Four column washings (2 mL of 2M HCl each) were required to remove all Bi prior to elution of 211At. No bismuth was detected in solution from the 4th washing in any of the elutions studied. Low (< 6%) recovery of 211At from the AG1×8 columns was obtained using the conditions studied. Good (60–79%) recovery of 211At was obtained from PEG-coated resin using 15M NH4OH. Isolation of the 211At from NH4OH solution was accomplished by distillation. In an initial study 211At distilled before obtaining a dry residue. However, later studies demonstrated that addi-tion of NaOH prior to distillation kept the 211At in the distilling flask. These studies demonstrated that PEG-coated columns could be used to isolate 211At from HNO3-dissolved bismuth targets with good non-optimized (~60%) overall recovery yields. The studies are continuing with optimization of elu-tion conditions and automation of the process.

Introduction Since 2005, we produce, at academic scale in Orleans, 211At for needs of chemistry and physicians teams of Nantes in research project of alpha immunotherapy. Between 2005 and 2014, several modifications were been made on preparation of target, targetry and radiation to protect personnel. Material and Methods The first target was a molten Bi metal onto a Cu support pre-treated with acid attack. The wished thickness (up to 100 µm) was obtained by mechanical treatment of target. The target is irradiated at 32MeV alpha particle beam for around 2 hours and then delivered by road transport to users. Only a measure of radiation dose was made to evaluate target activity. The second target we have used since 2010 is a electrodeposition of Bi (thickness of around 30 µm) onto AlN backing. We used a beam of 30.5 MeV for reaction 207Bi(α,n)211At (2 h with a current intensity of 2µA). Activity has measured with a detector Ge at 687 keV (γ-branching fraction = 0.26 %) before to be delivered. For all targets, beam energy on target was around 28.7 Mev in order not to produce too much 210At. Results and Conclusion 138 productions with the first target were delivered with an estimated activity of less than 100 MBq. Difficulties with wet extraction1, low yield of radiolabelling (metallic impurities and activation of copper resulting in 66Ga and 67Ga) made necessary to change process of extraction. With support of AlN, dry extraction was used with good yield (75–80 %) and without activation of support. Until today, 46 batchs were delivered with activity of 44 ± 12 MBq/µAh. Yield activity of 211At has been almost doubled compared to first target (25MBq/µAh). The dose burden to personnel was decreased with modification of targetry (outside of blockhouse of cyclotron, in a specific line beam to radionuclide production, cf. FIG. 1). In the case of 211At production, energy of reaction is of major impact. With our versatile accelerator (range of energy in alpha between 10 and 50 MeV) and a low thickness of metal, it’s easy to reach the right energy. This radionuclide production will be continued until ARRONAX, Nantes cyclotron, could take over from us for bigger activity of 211At.

Difficulties with reproducibility of isolation yields when distilling 211At from irradiated bismuth targets led us to use a “wet chemistry” approach for that process1. The wet chemistry approach has provided 211At isolation yields of ~ 78 % after decay and Bi attenuation corrections2. However, the use of diisopropyl ether (DIPE) in the separation process has made it difficult to reach our goal of automating the 211At isolation. Therefore, we have investigated the use of column materials to simplify the isolation of 211At and remove DIPE from the process. In this investigation we evaluated the use of a strong anion exchange resin (AG1×8), a strong cation exchange resin (AG MP-50) and a polyethylene glycol (PEG)-coated resin for separation of 211At from the bismuth target material. Anion and cation resins AG1×8 and AG MP-50 were obtained from commercial sources. A PEG-coated resin was prepared by reaction of the Merrifield resin with mPEG-OH 2000 in the pres-ence of tBuOK at 80 °C for 3 days, followed by drying under vacuum. Prior to use of the PEG resin, it was soaked in H2O. Resins (400–800 mg) were loaded into polypropylene columns (Applied Separations, Inc.). Column elution studies were conducted with and without reductants (0.75M FeSO4/1M H2SO4 or Na2S2O5) to determine their effect on capture of 211At. After target dissolution in HNO3 (and in most cases subse-quent removal of HNO3 by distillation and redis-solution of solid in 8M HCl), 211At solution was loaded onto the column, then the column was washed with 2M HCl or H2O to separate the Bi, and finally was eluted with strong base to remove the 211At. Initial studies were conducted with stable iodine to determine if reductants were effective in the presence of large amounts of bismuth ions. Studies with AG1×8 used 125I to determine if that radiohalogen could be captured and recovered from the column when eluting with boric acid buffers at pH 5.3, 8.0 or 10, or H2O at pH 7. Capture and recovery of 211At was evaluated under the same conditions. Further studies with AG1×8 involved eluting with 4M H2SO4. A limited study with AG MP-50 resin used 1M HCl as eluant. Studies with PEG-coated columns used 2M HCl, 4M HCl, 8M HCl, 16 M HNO3 and 8M HNO3 as initial (capture) eluants. Strong base (0.2, 1 or 12.5 M NaOH; 15M NH4OH) and 3 or 500 mM tetrabutylammonium bromide (TBAB) were evaluated for removal of 211At from the columns tested. The efficiency for capture of 211At on the AG1×8 column was high (99%) when loading with strong acid, but decreased when using 0.1–0.2M boric acid (69–91 %) buffer. Low 211At capture efficiencies were obtained with AG MP-50 col-umns (15–29%). High 211At capture efficiencies (96–100%) were obtained with PEG-coated resins when loading with 8M HCl or 8M HNO3, irre-spective of whether reductant was in the acid solution. Four column washings (2 mL of 2M HCl each) were required to remove all Bi prior to elution of 211At. No bismuth was detected in solution from the 4th washing in any of the elutions studied. Low (< 6%) recovery of 211At from the AG1×8 columns was obtained using the conditions studied. Good (60–79%) recovery of 211At was obtained from PEG-coated resin using 15M NH4OH. Isolation of the 211At from NH4OH solution was accomplished by distillation. In an initial study 211At distilled before obtaining a dry residue. However, later studies demonstrated that addi-tion of NaOH prior to distillation kept the 211At in the distilling flask. These studies demonstrated that PEG-coated columns could be used to isolate 211At from HNO3-dissolved bismuth targets with good non-optimized (~60%) overall recovery yields. The studies are continuing with optimization of elu-tion conditions and automation of the process.

Introduction Since 2005, we produce, at academic scale in Orleans, 211At for needs of chemistry and physicians teams of Nantes in research project of alpha immunotherapy. Between 2005 and 2014, several modifications were been made on preparation of target, targetry and radiation to protect personnel. Material and Methods The first target was a molten Bi metal onto a Cu support pre-treated with acid attack. The wished thickness (up to 100 µm) was obtained by mechanical treatment of target. The target is irradiated at 32MeV alpha particle beam for around 2 hours and then delivered by road transport to users. Only a measure of radiation dose was made to evaluate target activity. The second target we have used since 2010 is a electrodeposition of Bi (thickness of around 30 µm) onto AlN backing. We used a beam of 30.5 MeV for reaction 207Bi(α,n)211At (2 h with a current intensity of 2µA). Activity has measured with a detector Ge at 687 keV (γ-branching fraction = 0.26 %) before to be delivered. For all targets, beam energy on target was around 28.7 Mev in order not to produce too much 210At. Results and Conclusion 138 productions with the first target were delivered with an estimated activity of less than 100 MBq. Difficulties with wet extraction1, low yield of radiolabelling (metallic impurities and activation of copper resulting in 66Ga and 67Ga) made necessary to change process of extraction. With support of AlN, dry extraction was used with good yield (75–80 %) and without activation of support. Until today, 46 batchs were delivered with activity of 44 ± 12 MBq/µAh. Yield activity of 211At has been almost doubled compared to first target (25MBq/µAh). The dose burden to personnel was decreased with modification of targetry (outside of blockhouse of cyclotron, in a specific line beam to radionuclide production, cf. FIG. 1). In the case of 211At production, energy of reaction is of major impact. With our versatile accelerator (range of energy in alpha between 10 and 50 MeV) and a low thickness of metal, it’s easy to reach the right energy. This radionuclide production will be continued until ARRONAX, Nantes cyclotron, could take over from us for bigger activity of 211At.

<p>Conventional cancer treatment based on radiotherapy or chemotherapy affects all dividing cells. By directing the therapy specifically to the tumor cells, normal cells can be spared. Tumor targeting molecules carrying a cytotoxic moiety is then an attractive approach. </p><p>In this thesis, an affibody molecule with high affinity for the protein HER-2, that is strongly associated with aggressive forms of breast cancer, was selected. After radiolabeling with ¹²⁵I, the affibody molecule, in monovalent and bivalent form, was tested <i>in vitro</i> in HER-2 overexpressing tumor cells and in transplanted tumors in mice. </p><p>It was shown that the HER-2 targeting affibody molecule bound its target in a specific manner, both <i>in vitro</i> and <i>in vivo</i>. The small size of the affibody molecule resulted in fast clearance through the kidneys. An impressive tumor-to-blood ratio of 10 eight hours post injection was achieved and the tumors could easily be visualized in a gamma camera. </p><p>The biologic effects of the bivalent affibody molecule and a monovalent affinity maturated version was measured and compared with the effects of the monoclonal antibody trastuzumab. It was found that although all molecules target the same protein, the effects differed greatly.</p><p>The affibody molecule was also labeled with the alpha-emitting radionuclide ²¹¹At. Specific decrease in survival was seen in HER-2 overexpressing cells receiving the ²¹¹At labeled affibody molecule. The sensitivity to the treatment differed between cell lines, probably as a result of differences between the cell lines in internalization and nuclear size. The ²¹¹At labeled affibody molecules were also tested <i>in vivo</i>, where stability of the ²¹¹At label was a problem. To circumvent this problem, more stable conjugation chemistry was tested, as well as strategies to prevent uptake of released ²¹¹At by normal organs.</p><p>This thesis describes the selection and optimization of affibody molecules for medical use for the first time.</p>

Conventional cancer treatment based on radiotherapy or chemotherapy affects all dividing cells. By directing the therapy specifically to the tumor cells, normal cells can be spared. Tumor targeting molecules carrying a cytotoxic moiety is then an attractive approach. In this thesis, an affibody molecule with high affinity for the protein HER-2, that is strongly associated with aggressive forms of breast cancer, was selected. After radiolabeling with 125I, the affibody molecule, in monovalent and bivalent form, was tested in vitro in HER-2 overexpressing tumor cells and in transplanted tumors in mice. It was shown that the HER-2 targeting affibody molecule bound its target in a specific manner, both in vitro and in vivo. The small size of the affibody molecule resulted in fast clearance through the kidneys. An impressive tumor-to-blood ratio of 10 eight hours post injection was achieved and the tumors could easily be visualized in a gamma camera. The biologic effects of the bivalent affibody molecule and a monovalent affinity maturated version was measured and compared with the effects of the monoclonal antibody trastuzumab. It was found that although all molecules target the same protein, the effects differed greatly. The affibody molecule was also labeled with the alpha-emitting radionuclide 211At. Specific decrease in survival was seen in HER-2 overexpressing cells receiving the 211At labeled affibody molecule. The sensitivity to the treatment differed between cell lines, probably as a result of differences between the cell lines in internalization and nuclear size. The 211At labeled affibody molecules were also tested in vivo, where stability of the 211At label was a problem. To circumvent this problem, more stable conjugation chemistry was tested, as well as strategies to prevent uptake of released 211At by normal organs. This thesis describes the selection and optimization of affibody molecules for medical use for the first time.

Ionisierende Strahlung verursacht in Abhängigkeit von den strahlenphysikalischen Eigenschaften der Radionuklide Zellschäden unterschiedlicher Komplexität. An humanen Lymphozyten wurde untersucht, ob die biologische Wirksamkeit von Alpha- und Betastrahlung sowie der Einfluss von Doxorubicin der Qualität des Strahlenschadens zugewiesen werden kann. Die DNA-Schäden und deren Reparatur wurden mit zellbiologischen Methoden quantifiziert.

Ionisierende Strahlung verursacht in Abhängigkeit von den strahlenphysikalischen Eigenschaften der Radionuklide Zellschäden unterschiedlicher Komplexität. An humanen Lymphozyten wurde untersucht, ob die biologische Wirksamkeit von Alpha- und Betastrahlung sowie der Einfluss von Doxorubicin der Qualität des Strahlenschadens zugewiesen werden kann. Die DNA-Schäden und deren Reparatur wurden mit zellbiologischen Methoden quantifiziert.