0883-2897/91$3.00+ 0.00

Nucl. Med. Biol. Vol. 18, No. 4, pp. 383-387, 1991 ht. J. Radiat. Appl. Instrum. Parr B Printed in Great Britain

Pergamon Press plc

Potential Artifact for the Increase of Tumor Associated Antigens in Serum SamPles from Patients Injected with Monoclonal Antibodies I

P. FERRONI’*,

D. E. MILENIC’, M. ROSELLI’, J. A. CARRASQUILL02, A. RAUBITSCHEK3, J. SCHLOM’t and D. COLCHER’$

‘Laboratory of Tumor Immunology and Biology, 2Department of Nuclear Medicine, Clinical Center, ACRF and ‘Radiation Oncology Branch, National Cancer Institute, National Institutes of Health, 9ooo Rockville Pike, Bethesda, MD 20892, U.S.A. The administration of [13’1]B72.3 MAb for both diagnostic and therapeutic purposes is followed by the development of human anti-mouse antibodies (HAMA) in approx. 70% of patients receiving i.v. doses of MAb above 1 mg. The presence of HAMA in circulation interferes in the detection of the tumorassociated glycoprotein TAG-72 using the CA 72-4 RIA. We have reported elsewhere (Ferroni er al., 1990) that heat treatment at !+O”C,pH 6.5, for 15 min is capable of precipitating and/or inactivating HAMA without causing any significant loss in antigen recovery. Furthermore, the removal of the interfering activity leads to a more correct evaluation of TAG-72 serum levels in HAMA-positive serum samples and to a continued use of TAG-72 in the follow-up of patients undergoing MAb-based clinical protocols.

Introduction Serum samples

from carcinoma monoclonal antibodies

patients,

who have

(MAbs) for both diagnosis and therapy purposes, may exhibit large fluctuations in levels of tumor-associated antigens (TAAs) as measured using commercial radioimmunoassays (RIAs) (Primus et al., 1988; Morton et al., 1988). These levels are often unrelated to the clinical course of the disease. It is well known that the injection of murine MAbs may cause the development of a host immune response represented by the formation of human anti-mouse antibodies (HAMA) (Shroff et al., 1985; Shawler et al., 1985; Reynolds et al., 1989). Nevertheless, the presence of heterophilic antibodies (i.e. autoantibodies or antibodies against proteins of various animal origin) may be found in patients who never underwent clinical procedures using MAbs (Bock et al., 1985; Boscato and Stuart, 1988; McCarthy et al., 1988). The use of pharmaceuticals of animal origin may account for the generation of such heterophilic antibodies cross-reacting with murine MAbs (Dahlmann and Bridlingmaier, 1989); alternatively, received

*Pre.sent address: Department of Experimental Medicine, University of Rome “La Sapienza”, Viale Regina Elena 324, 00161 Rome, Italy. tAuthor for correspondence. fPresent address: Denartment of PatholoavlMicrobiolonv. University of Nebraska Medical Ceny&, Omaha, ci 68198, U.S.A.

they might be caused by autoantibodies, rheumatoid factor, etc. (McCarthy et al., 1988). The presence

of elevated levels of intefering antibodies in serum samples may also be responsible for the creation of a bridge between the two murine MAbs used as reagents in many commercial RIAs, thus resulting in false positive levels of the antigen being evaluated. The CA 72-4 RIA (Centocor, Malvern, Pa) (Gero et al., 1989) is a radioimmunoassay that uses MAbs B72.3 (Colcher et al., 1981) and CC49 (Muraro et al., 1988) to detect the presence of a novel tumor marker, TAG-72 (Johnson et al., 1986), in serum samples from patients with adenocarcinoma. Using this assay approx. 55% of patients with advanced gastrointestinal malignancies showed serum TAG-72 levels greater than an arbitrary cut off limit (6U/mL). Thirty-six percent of patients with adenocarcinoma of the lung, and 53% with ovarian cancer (stage IV) also showed elevated levels of TAG-72. Only 3.5% of normal sera obtained from blood donors had TAG-72 levels greater than 6 U/mL. Since tumor markers are being used in the followup of carcinoma patients (Go and Zamcheck, 1982; Martin et al., 1985; Guadagni et al., 1990), it becomes important to correctly evaluate the RIA results in patients undergoing MAb-based immunodiagnostic and/or immunotherapeutic protocols and therefore to identify any artifactual increase or decrease of the TAA under investigation. 383

384

P. l%3tRONI

et 01.

l 2cin

0

100

30

150

Days

Ot.,....., 30

loo

xx)

IOCO

3cco

10,CCQ

HAMA tker

Fig. 1. Correlation between the development of HAMA and fluctuations in reported TAA serum levels. Serum samples

from one patient with colon cancer (GH) were assayed at various time points following the administration of [t3’I]B72.3MAb. A, CA 125 serum levels; n , CA 19-9 serum levels; 0, TAG-72 serum levels; *HAMA titer. Radiolabeled MAb B72.3, a murine IgGt reactive with the TAG-72 antigen, has been administered for immunoscintigraphic studies to over 1000 patients with colorectal and ovarian carcinomas (Esteban et al., 1987; Colcher ei al., 1987; Carrasquillo et al., 1988; Tuttle et al., 1988; Maguire et al., 1989). Patients injected with MAb B72.3 have been shown to develop HAMA as early as l-2 weeks following MAb administration (Colcher et al., 1990). Approximately 70% of patients receiving i.v. doses ranging from 1 to 20 mg are positive for the presence of HAMA, with a much lower percentage (20%) at MAb doses below 1 mg (Colcher et al., 1990). We describe here the establishment of a method capable of abolishing HAMA interfere in the detection of TAG-72 using CA 72-4 RIA in serum samples from patients undergoing MAb-based clinical protocols. These studies have been described elsewhere in more detail (Ferroni et al., 1990).

HAMA Interference in Radioimmunoassays Serum samples were obtained from 9 patients (6 with colon cancer and 3 with ovarian cancer) entered into a therapeutic protocol using [13’I]B72.3 MAb. HAMA titers, as well as TAG-72, CEA, CA 125 and CA 19-9 tumor markers were evaluated in samples drawn before each administration of the radiolabeled MAb and at various intervals after the injection. Increases of TAG-72 levels were found in serum samples from 8 of 9 patients at different time points following the administration of (“‘I]B72.3 MAb. The evaluation of the other tumor markers showed similar results. These increases were found to be coincident with the development of a positive titer of HAMA, thus suggesting a possible correlation between the presence of elevated TAA serum levels and the presence of a positive HAMA titer. Figure 1 shows the development of HAMA in a patient with colon cancer following the therapeutic injection of [“‘I]B72.3. The presence of HAMA is closely paral-

Fig. 2. Interference of anti-mouse immunoglobulins in the detection of TAG-72 using the CA 72-4 RIA. Primate anti-mouse antibodies were diluted at various titers in an ascitic fluid with 33 U/mL of TAG-72. -----, TAG-72 33 U/mL; -, presence of HAMA.

leled by the increase in reported TAG-72, CA 19-9 and CA 125 serum levels. No significant variations in CEA serum levels were observed before and after MAb administration. Figure 2 shows the extent of HAMA interference in the detection of TAG-72 using the CA 72-4 RIA. Primate anti-mouse antibodies from a cynomolgus monkey were diluted at various titers in an ascitic fluid from a patient with colon cancer, containing 33 U/mL of TAG-72. As shown, the presence of HAMA is responsible for an artifactual increase of reported antigen levels even at the lower titers usually found in patients’ sera.

Heat Treatment of Serum Samples Heat treatment at 90°C for 15 min was chosen for its simplicity, and the ability to process a large number of samples for routine assay. This choice was supported by the previous use of this treatment for CEA extraction (Rim et al., 1979) and in the elimination of HAMA interference in immunoassays for the detection of CEA (Primus et al., 1988). Major differences were observed however, in antigen recovery following a treatment at 90°C under different pH values. As shown in Table 1 the use of heat treatment at pH 5.0 resulted in considerable loss in the recovery of TAG-72, CA 125 and CA 19-9. In contrast, heat treatment at !Xl”C, pH 6.5, resulted in a recovery of TAG-72 of approx. 90%. The efficacy of this procedure has been confirmed by TAG-72 determinations on serum samples obtained from the 9 patients (Table 2). As shown for patient WJ, no substantial differences were found between the untreated and heat-treated HAMAnegative samples, thus confirming that heat treatment at 9O”C, pH 6.5, does not alter TAG-72 detection using CA 72-4 RIA. The results of heat treatment on serum samples obtained from patients with positive titers of circulating HAMA are also shown in Table 2. An increase of TAG-72 appears at various times after the

Interference by HAMA in the CA 72-4 RIA

385

Table 1. Effect of heat treatment on antigen levels in radioimmunoassays Percent of antigen recovery in RIAs Treatment

CEA

CA 72-4

CA 125

CA 19.9

pH 5.0 pH 5.0/9VC pH 6.5 DH 6.5/9o”C

90 12 107 87

105 57 101 90

Potential artifact for the increase of tumor associated antigens in serum samples from patients injected with monoclonal antibodies.

The administration of [131I]B72.3 MAb for both diagnostic and therapeutic purposes is followed by the development of human anti-mouse antibodies (HAMA...
491KB Sizes 0 Downloads 0 Views