EuropeanJournalof
Nuclear Medicine
Original article
Reduced technetium 99m labelled NCA-95/CEA-antibody uptake in liver due to gentle antibody reconstitution Technical
note
S.N. Re s k e a n d U. B u e l l Department of Nuclear Medicine, RWTH Aachen, Pauwelsstrasse, D-5100 Aachen, Federal Republic of Germany Received September 11, 1989 and in revised form November 2, 1989
Abstract. The influence of reconstituting a murine
monoclonal IgG1 antibody kit with pertechnetate technetium 99m on antibody distribution in the liver, spleen and sternal bone marrow of patients was examined. The 99mTc-labelled antibody used is directed against nonspecific cross-reacting antigen (NCA-95) and carcinoembryonic antigen (CEA) and has been successfully applied for imaging tissue inflammation and bone marrow scanning. Radioactivity uptake was determined in the liver, spleen, bone marrow and a precordial background region in a consecutive series of 25 patients, examined with an antibody preparation, routinely radiolabelled according to the manufacturer's recommendations and in 14 patients, in whom the antibody was reconstituted with special care, avoiding bubble formation and dropping of buffer into the antibody-containing vial. Gentle compared with routine antibody reconstitution caused a highly significant reduction of the antibody uptake in the liver, as determined by count densities, normalized to injected dose and acquisition time (13.2_+5.5 vs 20.1+6.0 cpm per pixel, 2 ± S D , P=0.008). The liver to background ratio was reduced from 3.4_ 1.4 to 1.9_ 0.5 (P < 0.001). Spleen, sternal bone marrow and precordial background count rates were not significantly affected. These results clearly demonstrate that gentle antibody reconstitution can decrease non-specific antibody uptake in the liver by 34% _+6.4% (f~_+SEM). Thus, scan quality is improved, and the potential deleterious camouflage of underlying structures is avoided. Key words." Monoclonal antigranulocyte a n t i b o d y -
NCA-95-specific antibody - Bone marrow scintigraphy Antibody liver uptake Eur J Nucl Med (1990) 17:38-41 Offprint requests to: S.N. Reske
Introduction
A technetium 99m murine monoclonal antibody directed against granulocytes in peripheral blood and mature granulopoetic marrow cells has been recently introduced for localization of tissue inflammation and bone marrow scanning in patients (Bosslet et al. 1985; Steinstraesser et al. 1988; Joseph et al. 1988; Reske et al. 1989). This antibody belongs to the IgG1 subclass and detects nonspecific cross-reacting antigen (NCA-95) exposed at the cell membrane of granulocytes and mature myelocytes (Wahren et al. 1980; Burtin et al. 1975), as well as carcinoembryonic antigen (CEA) (Bosslet et al. 1985). The 99mTc-labelled antibody (TcNCAA) has been successfully used for in vivo radiolabelling of granulocytes and thus imaging tissue inflammation (Joseph et al. 1988). Due to fast and stable binding of a considerable amount of intravenously injected antibody to myelocytes, TcNCAA provides an excellent radiopharmaceutical for bone marrow scanning (Reske et al. 1989). Following this approach, first results indicated that metastatic spread to the skeleton may be detected with a higher sensitivity than by bone scanning in patients with carcinoma of the breast and malignant lymphomas (Reske et al. 1989). A significant uptake of TcNCAA in the liver of patients was observed in our laboratory as well as by others (Joseph et al. 1988; Reske et al. 1989; Berberich et al. 1989). Occasional observations showed extremely increased TcNCAA uptake in the liver, when harsh reconstitution conditions of the lyophilized antibody kit were applied. Thus, TcNCAA binding to the liver was probably not exclusively mediated by the biological properties of the antibody itself but also to some extent by protein aggregation or denaturation produced during reconstitution of the kit. We therefore examined whether reconstituting the antibody kit with special care would
© Springer-Verlag 1990
39 alter its biodistribution as assessed by u p t a k e o f the labelled a n t i b o d y in the liver, spleen and sternal b o n e marr o w o f patients.
Statistics. Differences between means were tested for significance
Material and methods
Results
Antibody. TcNCAA is a murine monoclonal isotype IgG1 antibody
In all patients, there was a fast and intense a n t i b o d y u p t a k e in the h a e m a t o p o e t i c b o n e m a r r o w . Just 3 h after T c N C A A injection, the b o n e m a r r o w was delineated with high contrast. M a r r o w to b a c k g r o u n d ratios, assessed f r o m the sternum to precordial R O I , was 3.1 _+ 1.14 in series 1 patients and 3.48_+1.04 in series 2 patients. This difference was statistically n o t significant (P=0.33). In all series 1 patients, a m o d e r a t e a n t i b o d y u p t a k e in the liver was observed; in contrast, series 2 patients h a d only faint T c N C A A u p t a k e in the liver. The u p t a k e was 20.1 ___6.0 counts/pixel vs 13.2__ 5.5, respectively, and was thus reduced by 34% +__6.4% (2 ± S E M ) by the gentle reconstitution procedure. A n t i b o d y u p t a k e in the liver o f patients in b o t h series was highly significantly different ( P = 0.008) (Fig. 1). As a result, liver to spleen and liver to b a c k g r o u n d ratios were significantly increased after routine a n t i b o d y reconstitution (Fig. 2). T c N C A A u p t a k e as assessed by c o u n t densities in the spleen a n d b o n e m a r r o w o f patients in b o t h g r o u p s was n o t significantly c h a n g e d by the kit reconstitution conditions (Fig. 1). C o u n t densities registered over the spleen were 12.8 _+4.8 cpm/pixel for the carefully reconstituted a n t i b o d y vs 1 4 . 6 + 8 . 6 for the routine preparation. The respective values for the b o n e m a r r o w were 23.7_+ 10.0 vs 20.0-t-7.5 cpm/pixel and 7.1 _+2.8 vs 6 . 6 + 2.1 for the precordial b a c k g r o u n d region. These differences were n o t significant. Accordingly, spleen and b o n e m a r r o w to b a c k g r o u n d ratios showed no significant differences between the two a n t i b o d y reconstitution procedures (Fig. 2).
(Szintimmun Granulozyt, Hoechst AB, Frankfurt, FRG). The 99mTc-labelled antibody is of excellent in vivo stability, sustained immunoreactivity above 80% of the native material and has an affinity constant of 2 x 109 1/mole (Steinstraesser et al. 1988). Lyophilized antibody (1 mg) is supplied as a kit for radiolabelling with Tc 99m as described by Schwartz and colleagues (Schwarz and Steinstraesser 1987). Briefly, 2.7mg 1,1,3,3-propanetetraphosphonic acid and 0.12 mg tin-(II)-chloride are dissolved in 5 ml saline under an atmosphere of nitrogen. Then 1 ml of this solution is transferred into the lyophilized antibody-containing vial. Care was taken not to shake the antibody solution-containing vial in order to avoid foam formation. Dropping of the reduced tin-containing solution into the lyophilized antibody containing vial was allowed. After a 10 min incubation time, 1.2 GBq pertechnetate Tc 99m, eluted from a 5.5 GBq 99Mo/99mTc generator (Elumatic III, CIS, Gif-surYvette, France) was mixed with the dissolved protein. This procedure is called "routine reconstitution" throughout the text. In order to minimize protein denaturation, the kit was reconstituted in a series of 14 examinations with special care, avoiding bubble formation or dropping of the tin- or pertechnetate Tc 99mcontaining solutions into the antibody-containing vial. This procedure is called '° gentle reconstitution". Contact of the solutions with oxygen was avoided during both reconstitution and labelling procedures. First eluates after generator delivery were not used. Patients received 100-250 gg of protein tagged with 300 ± 10 (2_+ 1 SD) MBq Tc 99m i.v. No acute or delayed adverse reactions from the antibody injection were observed. Patients. Antibody uptake in the liver of 25 consecutively examined patients (series 1), injected with TcNCAA labelled by the routine procedure, was compared with that of 14 patients (series 2), in whom the antibody was reconstituted by the gentle procedure. Patients were imaged 3-4 h after TcNCAA injection by means of a LFOV gamma camera, connected to a minicomputer and equipped with a LEAP collimator. Spot views of the axial and proximal appendicular skeleton were recorded on X-ray film (Cronex MRS 31 DDS 100, DuPont, Bad Homburg) without blood pool or tissue background subtraction. Some 500 kilocounts were acquired per image of axial, bone marrow-containing skeleton and 100-200 kilocounts for the extremities. Quantitative analysis was performed from anterior spot views, showing most parts of the liver, spleen, sternum and precordial left ventricular region. Liver, spleen, bone marrow uptake and background count densities, normalized to I min acquisition time, were determined in rectangular regions of interest (ROI), placed inside the scintigraphically defined tissue areas. The precordial ROI was set beside the left caudal margin of the sternum and regarded as the blood pool and non-target tissue background ROI. The ROI size was 90-110 pixels (range), containing 4500-50000 cpm. Liver to spleen, to sternum and to background ratios were determined and used as indices for relative radioactivity distribution between the tissues analysed. These ratios and tissue count densities were compared in both patient groups.
by the t-test for independent samples. A difference was regarded as significant for P
Nuclear Medicine
Original article
Reduced technetium 99m labelled NCA-95/CEA-antibody uptake in liver due to gentle antibody reconstitution Technical
note
S.N. Re s k e a n d U. B u e l l Department of Nuclear Medicine, RWTH Aachen, Pauwelsstrasse, D-5100 Aachen, Federal Republic of Germany Received September 11, 1989 and in revised form November 2, 1989
Abstract. The influence of reconstituting a murine
monoclonal IgG1 antibody kit with pertechnetate technetium 99m on antibody distribution in the liver, spleen and sternal bone marrow of patients was examined. The 99mTc-labelled antibody used is directed against nonspecific cross-reacting antigen (NCA-95) and carcinoembryonic antigen (CEA) and has been successfully applied for imaging tissue inflammation and bone marrow scanning. Radioactivity uptake was determined in the liver, spleen, bone marrow and a precordial background region in a consecutive series of 25 patients, examined with an antibody preparation, routinely radiolabelled according to the manufacturer's recommendations and in 14 patients, in whom the antibody was reconstituted with special care, avoiding bubble formation and dropping of buffer into the antibody-containing vial. Gentle compared with routine antibody reconstitution caused a highly significant reduction of the antibody uptake in the liver, as determined by count densities, normalized to injected dose and acquisition time (13.2_+5.5 vs 20.1+6.0 cpm per pixel, 2 ± S D , P=0.008). The liver to background ratio was reduced from 3.4_ 1.4 to 1.9_ 0.5 (P < 0.001). Spleen, sternal bone marrow and precordial background count rates were not significantly affected. These results clearly demonstrate that gentle antibody reconstitution can decrease non-specific antibody uptake in the liver by 34% _+6.4% (f~_+SEM). Thus, scan quality is improved, and the potential deleterious camouflage of underlying structures is avoided. Key words." Monoclonal antigranulocyte a n t i b o d y -
NCA-95-specific antibody - Bone marrow scintigraphy Antibody liver uptake Eur J Nucl Med (1990) 17:38-41 Offprint requests to: S.N. Reske
Introduction
A technetium 99m murine monoclonal antibody directed against granulocytes in peripheral blood and mature granulopoetic marrow cells has been recently introduced for localization of tissue inflammation and bone marrow scanning in patients (Bosslet et al. 1985; Steinstraesser et al. 1988; Joseph et al. 1988; Reske et al. 1989). This antibody belongs to the IgG1 subclass and detects nonspecific cross-reacting antigen (NCA-95) exposed at the cell membrane of granulocytes and mature myelocytes (Wahren et al. 1980; Burtin et al. 1975), as well as carcinoembryonic antigen (CEA) (Bosslet et al. 1985). The 99mTc-labelled antibody (TcNCAA) has been successfully used for in vivo radiolabelling of granulocytes and thus imaging tissue inflammation (Joseph et al. 1988). Due to fast and stable binding of a considerable amount of intravenously injected antibody to myelocytes, TcNCAA provides an excellent radiopharmaceutical for bone marrow scanning (Reske et al. 1989). Following this approach, first results indicated that metastatic spread to the skeleton may be detected with a higher sensitivity than by bone scanning in patients with carcinoma of the breast and malignant lymphomas (Reske et al. 1989). A significant uptake of TcNCAA in the liver of patients was observed in our laboratory as well as by others (Joseph et al. 1988; Reske et al. 1989; Berberich et al. 1989). Occasional observations showed extremely increased TcNCAA uptake in the liver, when harsh reconstitution conditions of the lyophilized antibody kit were applied. Thus, TcNCAA binding to the liver was probably not exclusively mediated by the biological properties of the antibody itself but also to some extent by protein aggregation or denaturation produced during reconstitution of the kit. We therefore examined whether reconstituting the antibody kit with special care would
© Springer-Verlag 1990
39 alter its biodistribution as assessed by u p t a k e o f the labelled a n t i b o d y in the liver, spleen and sternal b o n e marr o w o f patients.
Statistics. Differences between means were tested for significance
Material and methods
Results
Antibody. TcNCAA is a murine monoclonal isotype IgG1 antibody
In all patients, there was a fast and intense a n t i b o d y u p t a k e in the h a e m a t o p o e t i c b o n e m a r r o w . Just 3 h after T c N C A A injection, the b o n e m a r r o w was delineated with high contrast. M a r r o w to b a c k g r o u n d ratios, assessed f r o m the sternum to precordial R O I , was 3.1 _+ 1.14 in series 1 patients and 3.48_+1.04 in series 2 patients. This difference was statistically n o t significant (P=0.33). In all series 1 patients, a m o d e r a t e a n t i b o d y u p t a k e in the liver was observed; in contrast, series 2 patients h a d only faint T c N C A A u p t a k e in the liver. The u p t a k e was 20.1 ___6.0 counts/pixel vs 13.2__ 5.5, respectively, and was thus reduced by 34% +__6.4% (2 ± S E M ) by the gentle reconstitution procedure. A n t i b o d y u p t a k e in the liver o f patients in b o t h series was highly significantly different ( P = 0.008) (Fig. 1). As a result, liver to spleen and liver to b a c k g r o u n d ratios were significantly increased after routine a n t i b o d y reconstitution (Fig. 2). T c N C A A u p t a k e as assessed by c o u n t densities in the spleen a n d b o n e m a r r o w o f patients in b o t h g r o u p s was n o t significantly c h a n g e d by the kit reconstitution conditions (Fig. 1). C o u n t densities registered over the spleen were 12.8 _+4.8 cpm/pixel for the carefully reconstituted a n t i b o d y vs 1 4 . 6 + 8 . 6 for the routine preparation. The respective values for the b o n e m a r r o w were 23.7_+ 10.0 vs 20.0-t-7.5 cpm/pixel and 7.1 _+2.8 vs 6 . 6 + 2.1 for the precordial b a c k g r o u n d region. These differences were n o t significant. Accordingly, spleen and b o n e m a r r o w to b a c k g r o u n d ratios showed no significant differences between the two a n t i b o d y reconstitution procedures (Fig. 2).
(Szintimmun Granulozyt, Hoechst AB, Frankfurt, FRG). The 99mTc-labelled antibody is of excellent in vivo stability, sustained immunoreactivity above 80% of the native material and has an affinity constant of 2 x 109 1/mole (Steinstraesser et al. 1988). Lyophilized antibody (1 mg) is supplied as a kit for radiolabelling with Tc 99m as described by Schwartz and colleagues (Schwarz and Steinstraesser 1987). Briefly, 2.7mg 1,1,3,3-propanetetraphosphonic acid and 0.12 mg tin-(II)-chloride are dissolved in 5 ml saline under an atmosphere of nitrogen. Then 1 ml of this solution is transferred into the lyophilized antibody-containing vial. Care was taken not to shake the antibody solution-containing vial in order to avoid foam formation. Dropping of the reduced tin-containing solution into the lyophilized antibody containing vial was allowed. After a 10 min incubation time, 1.2 GBq pertechnetate Tc 99m, eluted from a 5.5 GBq 99Mo/99mTc generator (Elumatic III, CIS, Gif-surYvette, France) was mixed with the dissolved protein. This procedure is called "routine reconstitution" throughout the text. In order to minimize protein denaturation, the kit was reconstituted in a series of 14 examinations with special care, avoiding bubble formation or dropping of the tin- or pertechnetate Tc 99mcontaining solutions into the antibody-containing vial. This procedure is called '° gentle reconstitution". Contact of the solutions with oxygen was avoided during both reconstitution and labelling procedures. First eluates after generator delivery were not used. Patients received 100-250 gg of protein tagged with 300 ± 10 (2_+ 1 SD) MBq Tc 99m i.v. No acute or delayed adverse reactions from the antibody injection were observed. Patients. Antibody uptake in the liver of 25 consecutively examined patients (series 1), injected with TcNCAA labelled by the routine procedure, was compared with that of 14 patients (series 2), in whom the antibody was reconstituted by the gentle procedure. Patients were imaged 3-4 h after TcNCAA injection by means of a LFOV gamma camera, connected to a minicomputer and equipped with a LEAP collimator. Spot views of the axial and proximal appendicular skeleton were recorded on X-ray film (Cronex MRS 31 DDS 100, DuPont, Bad Homburg) without blood pool or tissue background subtraction. Some 500 kilocounts were acquired per image of axial, bone marrow-containing skeleton and 100-200 kilocounts for the extremities. Quantitative analysis was performed from anterior spot views, showing most parts of the liver, spleen, sternum and precordial left ventricular region. Liver, spleen, bone marrow uptake and background count densities, normalized to I min acquisition time, were determined in rectangular regions of interest (ROI), placed inside the scintigraphically defined tissue areas. The precordial ROI was set beside the left caudal margin of the sternum and regarded as the blood pool and non-target tissue background ROI. The ROI size was 90-110 pixels (range), containing 4500-50000 cpm. Liver to spleen, to sternum and to background ratios were determined and used as indices for relative radioactivity distribution between the tissues analysed. These ratios and tissue count densities were compared in both patient groups.
by the t-test for independent samples. A difference was regarded as significant for P
