EuropeanJournalof
Nuclear Medicine
Original article
Detection of colorectal carcinomas by intraoperative RIS in addition to preoperative RIS: surgical and immunohistochemical findings M. Reuter 1, R. Montz 1, K. de Heer 2, H. Sch&fer 3, R. K l a p d o r 4, K. D e s l e r 1 and H.W. S c h r e i b e r 2 Departments of ' Nuclear Medicine, 2 General Surgery, 3 Pathology and 4 Internal Medicine, University Hospital, Martinistrasse 52, W-2000 Hamburg 20, Federal Republic of Germany
Abstract. The immunoscintigraphic detection of tumour foci < 1 cm in diameter fails even with single photon emission tomography (SPET) owing to low contrast against background activity. In an attempt to improve detection of macroscopically invisible tumour spread, intraoperative scintimetry (IOSM) with a hand-held gamma-probe was performed in addition to SPET 24-30 and 41~48 h after injection of the technetium-99m carcinoembryonic antigen (CEA MoA) on 12 patients with colorectal carcinoma and 3 patients with different neoplastic and inflammatory diseases. Turnout specimens were measured in vitro in a gamma well counter. For comparison, the presence and amount of CEA in the tumour cells were evaluated immunohistochemically. After modification, the gamma-probe originally designed for iodine-131 was 20 times more sensitive; activities of 99mTc located close to the collimator hole were measured with absolute sensitivity of 100 cps=2.5 kBq 99mTC. The unfavourably high background activity affected both the in vitro and in vivo analysis: SPET results had been considered positive in 8 of 15 cases. In vitro tumour/non-tumour (t/n0 ratios > 2.0 were found in 4 cases. In vivo IOSM resulted in t/nt ratios >2.0 in only 3 patients. In most cases, there was no coincidence of elevated t/nt ratios from the different methods. A correlation was derived between positive immunoscintimetric in vitro findings and immunohistochemically proven interstitial localization of CEA in tumor cells. In conclusion, the measurement technique of IOSM seems adequate, but clinical success will depend on a more specific enrichment of MoA in tumour tissue. Future in vivo studies should be performed mainly in cases with a positive immunohistochemistry (interstitial CEA localization) result for the primary tumour.
Key words: Immunoscintigraphy - Intraoperative immunoscintimetry - Gamma probe - Technetium-99m carcinoembryonic antigen-specific antibody - Colorectal neoplasm Offprint requests to: M. Reuter
Eur J Nucl Med (1992) 19:102-109
Introduction The main indication for immunoscintigraphy using radiolabelled monoclonal carcinoembryonic antigen (CEA)-specific antibodies (MoAbs) is the early detection of recurrences of colorectal carcinoma in patients with steadily increasing CEA-/CA 19-9 serum levels and negative or questionable findings on computed tomography (CT) study (Bares et al. 1989; Baum et al. 1987; Kroiss et al. 1989; Lind et al. 1989). However, it is well-known that the immunoscintigraphic visualization of small turnout foci ( < 1 cm) is generally unsuccessful even by means of single photon emission tomography (SPET) owing to the low contrast compared with the non-specific background activity (Abdel-Nabi et al. 1988; Kramer et al. 1988; Lind et al. 1989; Schlag et al. 1987; Siccardi et al. 1989). Intraoperative immunoscintimetric examination with a hand-held gamma-detector probe in addition to preoperative external immunoscintigraphy may give new perspectives in tumour diagnostics with MoAbs. Because of the physical characteristics of the applied nuclide iodine-125 intraoperative studies have so far been conducted without prior external immunoscintigraphy (Bertoglio et al. 1989; Martin et al. 1988; Nieroda et al. 1989; Sardi etal. 1989; Sickle-Santanello etal. 1987; Tuttle et al. 1988). Since the development of an appropriate method for labelling (Schwarz and Steinstrfisser 1987), immunoscintigraphy with technetium-99m-labelled antibodies is possible. Initial clinical experiences with this ahnost ideal radionuclide promise favourable results in immunoscintigraphic tumour diagnostics (Bares et al. 1989; Baum et al. 1989; Kroiss et al. 1989; Lind et al. 1989). In an effort to improve detection of macroscopically invisible turnout spread, we combined intraoperative immunoscintimetric examination and preoperative SPETimmunoscintigraphy using 99mTc-CEA-specific MoAb
© Springer-Verlag 1992
103
in patients with colorectal carcinoma. In the present paper we describe our initial experiences obtained by investigation of 15 patients.
Table 1. Study group characteristics Patient Sex, Proven diagnosis, (No.) age preoperative therapy (years)
TNM
0.5
Patients and methods
1
M 62
Rectum Ca RTX 25 Gy
Patients. The study group consisted of 7 female and 8 male patients
2
M 60
Met. sigma Ca Chemotherapy
3
M 75
Colon ascendens Ca Untreated
T3 N2 M1
7.1
4
F 74
Rectum Ca RTX 25 Gy
T3 NO M0
0.5
5
F 63
Villous adenoma colon ascendens, recurrent rectum Ca Untreated
6
F 74
Sigma Ca Untreated
T3 N0 N0
7
F 75
Caecum Ca Untreated
T3 NO M0
0.5
8
M 66
Secondary neoplasm of colon (left flexura) Untreated
T3 N2 M0
0.5
9
F 69
Colon transversum Ca Untreated
T2 N0 M0
2.5
10
M 65
Colon Ca (left flexura) Untreated
11
M 65
Rectum Ca RTX 25 Gy
T2 N2 M0
0.5
12
F 70
Met. rectum Ca Two operations, RTX 50 Gy, chemotherapy
T3 NO MI
48.5
13
F 48
Infiltrating uterus Ca RTX
T4 N I M0
No data
14
M 50
Pancreas Ca Operation
T3 N1 M0
No data
15
M 44
Chronic pancreatitis
ranging in age from 44 to 75 years. Most of them suffered from colorectal carcinoma (n = 12), one patient each was ill with chronic pancreatitis and carcinoma of the pancreas. In every case diagnosis was proven by surgical and histological findings. One patient with previously proven and treated carcinoma of the uterus (patient no. 13) was free of turnout during reoperation. Seven patients were untreated prior to examination, whereas 4 were preoperatively irradiated, and 1 with metastatic sigma carcinoma had previously received chemotherapy. One patient with metastasizing rectum carcinoma had undergone 2 operations, irradiation, and chemotherapy preoperatively. A patient with pancreas carcinoma was reoperated. Table 1 lists the sex, age, diagnoses established by surgical and pathological findings, preoperative therapy, staging, and CEA values of our patients. Half of the tumours were in an advanced stage, with penetration of the colon wail, classified as T3 in accordance with U I C C classification. Seven of 15 patients showed normal levels of plasma CEA concentration. Five patients had marginal or moderately elevated CEA values, and a high positive plasma CEA was detectable in only 1 patient. The CEA values of 2 patients are not documented.
Radiolabelled drug. The specific intact monoclonal antibody BW 431/26 from Behring which reacts with the membrane CEA (Bosslet et al. 1988) was labelled with pertechnetate Tc99m with the aid of a purchased kit (Scintimun CEA-Tc 99m, Behring AG). The radiolabelled drug was administered i.v. over 10 min in a dosage of 20 MBq activity per kg body weight 1 or 2 days prior to the operation. Intolerance reactions were not observed during or following application of the murine antibody. External scintigraphy. The immunoscintigraphy was performed using SPET immediately and 4-6 h after infusion as well as on the day of operation 20-24 h p.i. (n--9) or 38-42h p.i. (n=6). For further technical detail see (Montz et al. 1986). The reconstructed transverse, coronal, and sagittal layers were judged with knowledge of the complete preoperative findings.
In vitro analysis. In an effort to calculate the percentage tracer uptake per gram tumour tissue, we measured tissue specimens from the tumour, normal colon, and suspected metastatic tissue. The analysis was performed in vitro in a gamma well counter.
Intraoperative scintimetry. Intraoperative immunoscintimetric measurements were performed 24-30 h (n = 9) or 42-47 h (n = 6) after tracer infusion. Measurements were made as follows : standard with cobalt-57, turnout(s) and surrounding tissue, unsuspicious tissue, suspected locoregional tumour spread and final standard. Intraoperative immunoscintimetry was performed using a modified hand-held gamma-detector probe consisting of a l " x 1" NaI(T1) crystal (Stratec Elektronik GmbH, Birkenfeld, F R G ) which was originally developed to measure gamma-energy of 131i. The probe was modified to such a degree that the sensitivity in detecting activities of 99mTC was considerably improved, while the
T1 N0 M0
CEA ~ (gg/1)
588
29
10
35
Nuclear Medicine
Original article
Detection of colorectal carcinomas by intraoperative RIS in addition to preoperative RIS: surgical and immunohistochemical findings M. Reuter 1, R. Montz 1, K. de Heer 2, H. Sch&fer 3, R. K l a p d o r 4, K. D e s l e r 1 and H.W. S c h r e i b e r 2 Departments of ' Nuclear Medicine, 2 General Surgery, 3 Pathology and 4 Internal Medicine, University Hospital, Martinistrasse 52, W-2000 Hamburg 20, Federal Republic of Germany
Abstract. The immunoscintigraphic detection of tumour foci < 1 cm in diameter fails even with single photon emission tomography (SPET) owing to low contrast against background activity. In an attempt to improve detection of macroscopically invisible tumour spread, intraoperative scintimetry (IOSM) with a hand-held gamma-probe was performed in addition to SPET 24-30 and 41~48 h after injection of the technetium-99m carcinoembryonic antigen (CEA MoA) on 12 patients with colorectal carcinoma and 3 patients with different neoplastic and inflammatory diseases. Turnout specimens were measured in vitro in a gamma well counter. For comparison, the presence and amount of CEA in the tumour cells were evaluated immunohistochemically. After modification, the gamma-probe originally designed for iodine-131 was 20 times more sensitive; activities of 99mTc located close to the collimator hole were measured with absolute sensitivity of 100 cps=2.5 kBq 99mTC. The unfavourably high background activity affected both the in vitro and in vivo analysis: SPET results had been considered positive in 8 of 15 cases. In vitro tumour/non-tumour (t/n0 ratios > 2.0 were found in 4 cases. In vivo IOSM resulted in t/nt ratios >2.0 in only 3 patients. In most cases, there was no coincidence of elevated t/nt ratios from the different methods. A correlation was derived between positive immunoscintimetric in vitro findings and immunohistochemically proven interstitial localization of CEA in tumor cells. In conclusion, the measurement technique of IOSM seems adequate, but clinical success will depend on a more specific enrichment of MoA in tumour tissue. Future in vivo studies should be performed mainly in cases with a positive immunohistochemistry (interstitial CEA localization) result for the primary tumour.
Key words: Immunoscintigraphy - Intraoperative immunoscintimetry - Gamma probe - Technetium-99m carcinoembryonic antigen-specific antibody - Colorectal neoplasm Offprint requests to: M. Reuter
Eur J Nucl Med (1992) 19:102-109
Introduction The main indication for immunoscintigraphy using radiolabelled monoclonal carcinoembryonic antigen (CEA)-specific antibodies (MoAbs) is the early detection of recurrences of colorectal carcinoma in patients with steadily increasing CEA-/CA 19-9 serum levels and negative or questionable findings on computed tomography (CT) study (Bares et al. 1989; Baum et al. 1987; Kroiss et al. 1989; Lind et al. 1989). However, it is well-known that the immunoscintigraphic visualization of small turnout foci ( < 1 cm) is generally unsuccessful even by means of single photon emission tomography (SPET) owing to the low contrast compared with the non-specific background activity (Abdel-Nabi et al. 1988; Kramer et al. 1988; Lind et al. 1989; Schlag et al. 1987; Siccardi et al. 1989). Intraoperative immunoscintimetric examination with a hand-held gamma-detector probe in addition to preoperative external immunoscintigraphy may give new perspectives in tumour diagnostics with MoAbs. Because of the physical characteristics of the applied nuclide iodine-125 intraoperative studies have so far been conducted without prior external immunoscintigraphy (Bertoglio et al. 1989; Martin et al. 1988; Nieroda et al. 1989; Sardi etal. 1989; Sickle-Santanello etal. 1987; Tuttle et al. 1988). Since the development of an appropriate method for labelling (Schwarz and Steinstrfisser 1987), immunoscintigraphy with technetium-99m-labelled antibodies is possible. Initial clinical experiences with this ahnost ideal radionuclide promise favourable results in immunoscintigraphic tumour diagnostics (Bares et al. 1989; Baum et al. 1989; Kroiss et al. 1989; Lind et al. 1989). In an effort to improve detection of macroscopically invisible turnout spread, we combined intraoperative immunoscintimetric examination and preoperative SPETimmunoscintigraphy using 99mTc-CEA-specific MoAb
© Springer-Verlag 1992
103
in patients with colorectal carcinoma. In the present paper we describe our initial experiences obtained by investigation of 15 patients.
Table 1. Study group characteristics Patient Sex, Proven diagnosis, (No.) age preoperative therapy (years)
TNM
0.5
Patients and methods
1
M 62
Rectum Ca RTX 25 Gy
Patients. The study group consisted of 7 female and 8 male patients
2
M 60
Met. sigma Ca Chemotherapy
3
M 75
Colon ascendens Ca Untreated
T3 N2 M1
7.1
4
F 74
Rectum Ca RTX 25 Gy
T3 NO M0
0.5
5
F 63
Villous adenoma colon ascendens, recurrent rectum Ca Untreated
6
F 74
Sigma Ca Untreated
T3 N0 N0
7
F 75
Caecum Ca Untreated
T3 NO M0
0.5
8
M 66
Secondary neoplasm of colon (left flexura) Untreated
T3 N2 M0
0.5
9
F 69
Colon transversum Ca Untreated
T2 N0 M0
2.5
10
M 65
Colon Ca (left flexura) Untreated
11
M 65
Rectum Ca RTX 25 Gy
T2 N2 M0
0.5
12
F 70
Met. rectum Ca Two operations, RTX 50 Gy, chemotherapy
T3 NO MI
48.5
13
F 48
Infiltrating uterus Ca RTX
T4 N I M0
No data
14
M 50
Pancreas Ca Operation
T3 N1 M0
No data
15
M 44
Chronic pancreatitis
ranging in age from 44 to 75 years. Most of them suffered from colorectal carcinoma (n = 12), one patient each was ill with chronic pancreatitis and carcinoma of the pancreas. In every case diagnosis was proven by surgical and histological findings. One patient with previously proven and treated carcinoma of the uterus (patient no. 13) was free of turnout during reoperation. Seven patients were untreated prior to examination, whereas 4 were preoperatively irradiated, and 1 with metastatic sigma carcinoma had previously received chemotherapy. One patient with metastasizing rectum carcinoma had undergone 2 operations, irradiation, and chemotherapy preoperatively. A patient with pancreas carcinoma was reoperated. Table 1 lists the sex, age, diagnoses established by surgical and pathological findings, preoperative therapy, staging, and CEA values of our patients. Half of the tumours were in an advanced stage, with penetration of the colon wail, classified as T3 in accordance with U I C C classification. Seven of 15 patients showed normal levels of plasma CEA concentration. Five patients had marginal or moderately elevated CEA values, and a high positive plasma CEA was detectable in only 1 patient. The CEA values of 2 patients are not documented.
Radiolabelled drug. The specific intact monoclonal antibody BW 431/26 from Behring which reacts with the membrane CEA (Bosslet et al. 1988) was labelled with pertechnetate Tc99m with the aid of a purchased kit (Scintimun CEA-Tc 99m, Behring AG). The radiolabelled drug was administered i.v. over 10 min in a dosage of 20 MBq activity per kg body weight 1 or 2 days prior to the operation. Intolerance reactions were not observed during or following application of the murine antibody. External scintigraphy. The immunoscintigraphy was performed using SPET immediately and 4-6 h after infusion as well as on the day of operation 20-24 h p.i. (n--9) or 38-42h p.i. (n=6). For further technical detail see (Montz et al. 1986). The reconstructed transverse, coronal, and sagittal layers were judged with knowledge of the complete preoperative findings.
In vitro analysis. In an effort to calculate the percentage tracer uptake per gram tumour tissue, we measured tissue specimens from the tumour, normal colon, and suspected metastatic tissue. The analysis was performed in vitro in a gamma well counter.
Intraoperative scintimetry. Intraoperative immunoscintimetric measurements were performed 24-30 h (n = 9) or 42-47 h (n = 6) after tracer infusion. Measurements were made as follows : standard with cobalt-57, turnout(s) and surrounding tissue, unsuspicious tissue, suspected locoregional tumour spread and final standard. Intraoperative immunoscintimetry was performed using a modified hand-held gamma-detector probe consisting of a l " x 1" NaI(T1) crystal (Stratec Elektronik GmbH, Birkenfeld, F R G ) which was originally developed to measure gamma-energy of 131i. The probe was modified to such a degree that the sensitivity in detecting activities of 99mTC was considerably improved, while the
T1 N0 M0
CEA ~ (gg/1)
588
29
10
35
