Identification and Measurement of Calcitonin Precursors in Serum of Patients with Malignant Diseases Pascale P. Ghillani, Philippe Motté, Frédéric Troalen, et al. Cancer Res 1989;49:6845-6851. Updated version E-mail alerts Reprints and Subscriptions Permissions Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/49/23/6845 Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at pubs@aacr.org. To request permission to re-use all or part of this article, contact the AACR Publications Department at permissions@aacr.org. Downloaded from cancerres.aacrjournals.org on October 26, 2014. © 1989 American Association for Cancer Research. (CANCER RESEARCH 49. 6845-6851. December I. 1989] Identification and Measurement of Calcitonin Precursors in Serum of Patients with Malignant Diseases' Pascale P. Chilian!, Philippe Motte, Frédéric Troalen, Annick Jullienne, Paule Gardet, Thierry Le Chevalier, Philippe Rougier, Martin Schlumberger, Claude Bohuon, and Dominique Bellet2 l'ailéd'Ã-mmunoehimie IP. P. (i., P. M., F. T., C. B., D. B.], ¡heDepartement de Médecine\ucleaire¡P. G., M. S./, the Seirice de MédecineB ¡T.L. C.J. and the Sen-ice de Gastroenterologie ¡P.R.], Institut (instare Roussy, rue Camille Desmoulins, 94805 l'illejuif Cedex, France, and L' 113 INSERM ¡A.J.] and l'A 163 CNRS, Hôpital ST Antoine, 27 rue de Chaligny, 75012 Paris, France ABSTRACT Previous studies have suggested that molecular species larger than the mature calcitonin «I ) are produced by tumors of different origin. In order to study these species, we developed a monoclonal immunoradiometric assay for calcitonin precursors (CT-pr). This assay was based on both monoclonal antibody KC01 directed to the 1-11 region of katacalcin and monoclonal antibody CT08 directed to the 11-17 portion of CT. The sensitivity of this monoclonal immunoradiometric assay for CT-pr was <100 pg/ml. Only one of 131 healthy subjects had CT-pr serum levels >100 pg/ml; this value was therefore selected as the standard serum value in healthy individuals. CT-pr was present in the serum of seven of ten patients with advanced renal failure and in that of 11 of 52 patients (40%) with benign liver disease but was undetectable in sera of patients with other benign diseases. The serum ( ' I -pr level was correlated with that of mature CT in patients with medullary carcinoma of the thyroid. In contrast, the serum CT-pr level was frequently elevated in the absence of a detectable CT level in patients with various malignant tumors and, particularly, in those with either tumors of the neuroendocrine system (60%) or hepatocellular carcinomas (62%). CT-pr was detected in tumor extract from a patient with a hepatocellular carcinoma. Moreover, hy bridization experiments with total RNA extracted from this tumor dem onstrated the presence of RNAs hybridizing with complementary DNA encoding for common region, calcitonin, and katacalcine sequences. These results show that CT precursors are excreted by numerous cancers and might well be useful biological markers for the follow-up of productive tumors. INTRODUCTION The hypocalcémiehormone CT3 is a 32-amino-acid-long polypeptide in its mature form which derives from the posttranslational processing of a larger precursor in thyroid C-cells (1). The complete sequence of this precursor, designated preprocalcitonin, has been deduced on the basis of the nucleotide sequence of cloned cDNAs encoding the peptide (2). Within the preprocalcitonin, CT is linked to the KC sequence and preceded by an 84-amino-acid-long N-terminal region that ter minates in a Lys-Arg cleavage signal. The biosynthesis of the hormone involves consecutive proteolytic events affecting pre procalcitonin: the signal peptide is removed as preprocalcitonin enters the endoplasmic reticulum to produce procalcitonin and the N-terminal region is then cleaved. The resulting 57-aminoacid-long polypeptide is composed of the CT sequence sepa rated from the KC sequence (21 residues) by the Gly-Lys-LysArg cleavage amidation site (3, 4). Finally, calcitonin is released by proteolysis, and the proline C-terminal residue is simulta neously amidated (5). Received 1/30/89; revised 7/21/89; accepted 9/1/89. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 17.14 solely to indicate this fact. ' This work was supported by a grant from "Association pour la Recherche sur le cancer," Villejuif and grant 88 D29 from Institut Gustave-Roussy. 2 To whom requests for reprints should be addressed. "*The abbreviations used are: CT. calcitonin; m-IRMA, monoclonal immuno radiometric assay; CT-pr. calcitonin precursor; mAb, monoclonal antibody; MCT. medullary carcinoma of the thyroid: cDNA. complementary DNA; KC. katacal cin; SCLC. small cell lung carcinoma; NSE. neuron-specific enolase. Neoplastic C-cells secrete large amounts of calcitonin in serum, and CT is used as a tumor-associated marker of MCT. Several groups have reported the heterogeneity of circulating forms of calcitonin and the presence of immunoreactive molec ular species of higher molecular weight than that of monomeric CT, but the precise nature of these forms has not been clearly determined (6). However, immunoprecipitable calcitonin-specific precursor molecules have been demonstrated following cell-free translation from either rat and human tissues or cell lines (7-9). These data suggested that the heterogeneity of immunoreactive forms of serum CT as well as the ectopie secretion of calcitonin described with radioimmunoassays based on polyclonal antibodies might be due to the presence of cir culating CT-pr. This study was aimed at both identifying the presence and determining the level of CT precursors in serum from healthy individuals and patients with either benign or malignant dis eases. We used a library of mAbs directed against distinct epitopes of the calcitonin or katacalcin molecule to develop two m-IRMAs for the specific identification and quantification of either mature calcitonin or biosynthetic CT precursors. We report here the simultaneous presence of both CT and CT-pr in sera of patients with MCT. In contrast, the presence of CTpr, usually in the absence of a detectable CT level, was found in sera of patients with malignant tumors of other origin. MATERIALS AND METHODS Subjects. We studied serum samples collected from a total of 876 individuals. The first time, 131 healthy individuals aged 20 to 60 yr and 30 pregnant women were evaluated as well as 101 patients with various benign diseases and 27 patients with MCT. The latter were divided into three different groups. A first group included untreated patients whose diagnosis of MCT was later confirmed (n = 5). A second group was composed of patients with no evidence of relapse and considered to be in complete remission (n = 19). This group was divided into two subgroups depending upon whether patients had (N+; n = 10) or did not have (N -, n = 9) lymph node involvement. Patients with recurrence of the disease (local neck recurrence or distant métastases; n = 7) were included in the third group. Patients whose sera were tested both before and after treatment or recurrence could be classified into two different groups. Moreover, pentagastrin stimulation tests were carried out in five patients with MCT and one healthy subject. After informed consent from each individual was obtained, the stimulation of CT secretion was performed by i.v. injection of pentagastrin (0.5 jig/ kg; Peptavlon; I.C.I. Pharma, Cergy, France) and serum samples were collected before and 2 and 5 min after infusion. Furthermore, serum samples from 587 patients with non-MCT ma lignant tumors of various origin were also studied. The presence of liver métastaseswas assessed by echographic examination. Finally, serial studies were carried out on samples collected from 3 subjects with SCLC. After drawing, blood samples were rapidly separated by centrifugation for 15 min at 1500 x g, and sera were rapidly stored frozen until analysis. After thawing, samples were heat inactivated (30 min at 56°C) to prevent enzymatic degradation of calcitonin in the serum (10) and 6845 Downloaded from cancerres.aacrjournals.org on October 26, 2014. © 1989 American Association for Cancer Research. SERUM CALCITONIN PRECURSOR MEASUREMENT assayed using different m-IRMAs. We have previously determined that heating the samples had no effect on immunoreactivity of either calcitonin or its precursors (data not shown). Production and Characterization of Monoclonal Antibodies to Calcitonin or to Katacalcin. The production and characterization of mono clonal antibodies directed to either mature CT or KC have already been described (11, 12). Briefly, these antibodies were obtained after fusion of splenocytes from mice immunized with either CT or KC, both conjugated to tetanus toxoid. Two monoclonal anti-CT antibodies designated CT07 and CT08 were directed against distinct and separate epitopes. (a) mAb CT07 (IgG2, A',sn= 0.9 x 10'°M~') recognized an epitope located in the 26-32 region of mature CT and did not bind to procalcitonin (11). It was previously shown that CT07 bound to the 26-32 sequence bearing the C-terminal carboxamide group present on mature CT and did not bind to a similar sequence bearing a C-terminal carboxyl group, (h) mAb CT08 (IgGl, A'„„ = 3.0 x 10'°M~') recognized an epitope present in the 11-17 portion of CT and which was also expressed on pcptidcs mimicking the sequence of procalcitonin. A monoclonal anti-katacalcin antibody identified as KC01 has also been characterized (12). mAbKCOl (IgGl, A'a.n= 1.5 x 109M~') recognized an epitope localized to the 1-11 region of katacalcin and also bound its epitope within the precursor molecule. The antibody binding sites of these three antibodies arc presented in Fig. 1. Development of Monoclonal Immunoradiometric Assays for Either Mature Calcitonin or Calcitonin Precursors. Monoclonal antibodies CT07, CT08, and KC01 were utilized to construct two different mono clonal immunoradiometric assays developed in a "one-step" (simulta neous) format. One assay based on CT07 and CT08 antibodies (mIRMA CT07-CT08) was used for the specific measurement of mature CT, while the other, based on KC01 and CT08 antibodies (m-IRMA KC01-CT08) was constructed for the specific quantification of CT-pr (12). Briefly, either CT07 or KC01 served as "capture antibody" and was linked to a solid phase support by incubating polystyrene beads (Polystyrene Plastic Balls, Chicago, IL) overnight at room temperature with a 500-fold dilution of the mAb-containing ascitic fluid in 0.1 M phosphate-buffered saline, pH 7.4. After extensive washing, the anti body-coated beads were incubated with sample or standard (200 n\), and mAb CT08( 100^1) was labeled with I25Iusing the lodogen method (13) and utilized as tracer (100,000 cpm). After overnight incubation at room temperature, the beads were washed with distilled water, and bound radioactivity was measured in a y scintillation counter. Synthetic calcitonin and a 47-amino-acid-long polypeptide analogous to the carboxyl-terminal part of procalcitonin (PTN47) were used as the standard in m-IRMA CT07-CT08 and m-IRMA KC01-CT08, respectively; the limit of detection of both assays was determined as previously described (12). Preparation of Tissue Extract. Liver tumor and normal liver tissues were placed in liquid nitrogen immediately after resection and stored at -80°C until time of extraction. The tissues were pulverized while still frozen and then homogenized with Polytron in phosphate-buffered saline containing 0.1% Aprotinin. After ccntrifugation (40,000 x g for 30 min at 4°C),the supernatant was collected and assayed by both mIRMAs. Affinity Chromatographies. Monoclonal antibody CT07 or KC01 (ascitic fluid) was coupled to CNBr-activated Sepharose 4B (Pharmacia, - PREPROCALCITONIN • PROCALCITONIN sequence « X N-terminal region Calcitonin > 4 Katacalcin 1 < CT07 DISEASE Serum sample from patient with MCT IRMA CT (CT07-CT08) IRMACT-pr(KC01-CT08) m-IRMA CT (A) m-IRMA CT-pr -»fa) m-IRMA CT-pr m-IRMA CT m-IRMA CT-pr Elution with acetic acid 2.5% (H) m-IRMACT m-IRMA CT-pr Fig. 2. Sequence of affinity Chromatographies followed by m-IRMAs for the characterization of CT and CT-pr in serum of a patient wilh MCT. Uppsala, Sweden) according to the manufacturer's instructions. Then, both gels were equilibrated with a Na2HPO4 0.05 Mstarting buffer (pH 8.5). Serum sample was incubated with the gel coupled with mAb KC01. After a rotating agitation for 18 h at 4°C,the gel was packed into a column; nonadsorbed material was eluted by extensive washing with distilled water, and fractions were collected every 5 min. The unbound fractions were then incubated with the gel coupled with mAb CT07. This latter gel was also stirred 18 h at 4°Cbefore packing into a column and collection of nonadsorbed material, as previously de scribed. Bound material was eluted from both columns with 2.5% acetic acid, and fractions were collected every 5 min. After neutralization, 200 ß\of all fractions diluted 1:1 in normal human serum were assayed by both CT07-CT08 and KC01-CT08 m-IRMAs. Fig. 2 shows the se quence of affinity Chromatographies followed by specific assays of bound and unbound fractions. Neuron-specific Enolase Measurement. NSE serum levels were deter mined using a commercial radioimmunoassay (Pharmacia, Uppsala, Sweden). This assay was performed according to the manufacturer's instructions. The normal range of NSE serum levels observed by this technique is <12.5 ng/ml. Extraction of Total RNA and Dot-Blot Hybridization. Total RNA was prepared by a modification of the method of Chomczynski and Sacchi (14). in which tissue was homogenized in guanidinium thiocyanatc and then extracted with phenol and chloroform-isoamyl alcohol mixture, before isopropanol precipitations. Purification of RNA was achieved by LiCl precipitation. RNA was quantified by measurement of absorbance at 260 nm. Purified total RNA denatured in formaldehyde for 15 min at 60°Cwas spotted onto GeneScreen (New England Nuclear, Boston, MA). The paper was air dried and then baked at 80°Cfor 2 h. The paper was then prehybridized in a buffer containing 50% formamide for 3 h, after which the nick-translated radiolabeled probe was added for 18 h at 42°C.In this experiment, we have used a cDNA encoding for common region, calcitonin and katacalcine sequences. After incubation, the paper was twice washed in a buffer containing 0.15 M NaCI, 0.015 M sodium citrate, and 0.1% sodium dodecyl sulfate at 55°Cfor 30 min each. The filter was then autoradiographed. T TT CTOa AND MALIGNANT KC01 Fig. I. Binding site of monoclonal antibodies to either the calcitonin or katacalcin molecule. MAbCTOS recognizes the CT [11-17] sequence. mAbCT07 recognizes the CT [26-32] amide sequence, and mAb KC01 recognizes the KC [1-11 ] sequence. RESULTS All serum samples were tested in two different assays. In one, we measured CT levels with the CT07-CT08 m-IRMA, which 6846 Downloaded from cancerres.aacrjournals.org on October 26, 2014. © 1989 American Association for Cancer Research. SERUM CALCITONIN PRECURSOR MEASUREMENT AND MALIGNANT DISEASE peak CT values ranged from 207 to 35,700 pg/ml and the peak/ basal ratios were higher than 2 (mean ±SEM = 14.4 ±7.4). Interestingly, the peak CT-pr values ranged from 189 to 91,800 pg/ml, but the peak/basal ratios of CT-pr were consistently lower than those of calcitonin (mean ±SEM 3.6 ±1.7) (Table 1). Using sequential affinity chromatography with either mAb KC01 or mAb CT07 linked to a CNBr-activated Sepharose column, we separated biosynthetic CT precursors from CT in serum of a given MCT patient whose initial serum levels of CT and CT-pr, determined by specific m-IRMAs, were 10,130 pg/ ml and 3,000 ng/ml, respectively. The serum was first laid onto the KC01 affinity column. After elution of unbound and then bound material, m-IRMAs performed on fractions demon strated that the unbound material showed CT immunoreactivity (Fig. 5, A and B). while the bound material collected after Fraction 10 displayed the peak of CT-pr immunoreactivity (Fig. 5, C and D). CT immunoreactive fractions (A) were then laid onto the CT07 affinity column. Assays performed on both unbound and bound fractions after elution demonstrated both the absence of significant CT-pr immunoreactivity in those fractions and the presence of a CT immunoreactive peak in the bound fractions collected after elution (Fig. 5, E to //). Measurement of CT-pr in Various Non-MCT Malignant Tu mors. After identification and measurement of biosynthetic calcitonin precursors in sera of patients with MCT, we extended the clinical study to non-MCT malignant tumors. First, we performed assays in sera of patients with tumors deriving from the neuroendocrine system, as does MCT. For this purpose, and in line with the Amine Precursor Uptake and Decarboxylation concept (16, 17), we tested sera from patients with either insulinoma, neuroblastoma, carcinoid tumors, malignant pheochromocytoma, small cell lung carcinoma, or nonspecific apudoma. These tumors are known for their neural crest origin and their potentiality for multiple peptide secretion (18). In particular, such malignancies have been described as producing neuropeptides (19, 20). We found that 7% of such patients had simultaneous elevation of CT and CT-pr levels, whereas 53% of patients had a detectable CT-pr level in the absence of a detectable CT value (Fig. 6). As 60% of patients with SCLC had elevated CT-pr serum levels, while only 6% had simulta neous elevation of the CT level, we studied whether or not the CT-pr level was increased in lung cancers of other histológica! is specific for mature calcitonin and displays a limit of detection of 10 pg/ml. In the other, we measured CT-pr levels with KC01CT08 m-IRMA, which is specific for biosynthetic CT precur sors and has a limit of detection of 100 pg/ml. In an attempt to estimate the normal range of serum CT-pr levels, we measured CT-pr values in serum samples collected from healthy individuals. We found that all but one subject had undetectable CT-pr serum levels (Fig. 3). The normal range was therefore defined as < 100 pg/ml. We also studied CT-pr levels in sera of patients with various benign diseases. Seven of 10 patients with advanced renal failure and 40% (21 of 52) of those with benign liver disease had elevated CT-pr serum values, while CT-pr was undetectable in sera of other patients (Fig. 3). It was noteworthy that 2 of 10 patients with renal failure had simultaneous elevation of both CT and CT-pr values. In con trast, CT was undetectable in sera of those with benign liver disease. Indeed, we had previously found an elevation of serum CT in 33% of patients with renal failure, but none in those with benign liver disease (15). Thus, the elevation of CT-pr in sera of patients with benign liver disease appears to be independent of secretion of mature calcitonin. Identification and Measurement of CT-pr in Patients with MCT. We measured both CT and CT-pr serum levels in MCT patients either untreated or with a residual tumor burden. These patients had CT levels in the range of 12 to 220,000 pg/ml. In those sera, CT-pr serum levels were consistently detectable and ranged from 112 pg/ml to 6.8 ng/m\. As shown by a regression curve, CT and CT-pr levels were correlated (r = 0.97), and it was estimated from this curve that CT-pr values were 11 times as high as CT values (Fig. 4). MCT patients in complete clinical remission, with no residual tumor burden and a serum CT level <10 pg/ml, had undetectable CT-pr levels, as did patients who underwent total thyroidectomy. Taken together, these results strongly suggested that neoplastic C-cells secrete calcitonin precursors as well as mature calcitonin. Furthermore, we investigated the effect of pentagastrin on CT and CT-pr secretion. For this purpose, sera from one healthy subject and from 5 MCT patients were evaluated before and 2 and 5 min after injection of pentagastrin. The healthy individual had undetectable CT and CT-pr levels during the stimulation test. In contrast, the five MCT patients had an elevated basal serum CT level (range, 19 to 1,180 pg/ml) and displayed a positive response to pentagastrin stimulation. The 70% Fig. 3. CT-pr serum levels in healthy subjects, pregnant women, and patients with benign diseases. •.10 cases; •.O, l case; O. patient with detectable CT. Numbers in parentheses. serum level of CT. The percentage represents the incidence of detectable CT-pr serum level in each group of patients. $ml # "3" J} .? r 6847 Downloaded from cancerres.aacrjournals.org on October 26, 2014. © 1989 American Association for Cancer Research. SERUM CALCITONIN PRECURSOR MEASUREMENT AND MALIGNANT DISEASE 75% 60% 60% 10' 10 01 3 OB»» Q. I Ãœ io, ...o"«' «119 .\ V IO1 CT (pg/ml) // • ' J Fig. 4. CT and CT-pr serum levels in patients with medullary carcinoma of the thyroid. Table I Effect ofpentagastrin stimulation on CT and CT-pr secretion Pentagastrin stimulation tests were carried out in one healthy subject (Patient A) and in five patients (B to F) as described in "Materials and Methods." CT-pr (pg/ml) at the follow ing times (min) CT (pg/ml) at the following times (min) Fig. 6. CT-pr serum levels in patients with tumors of the neuroendocrine system. •10 cases; •.O. +. 1 case: O. patient with detectable CT level (value indicated in parentheses); +. patient with liver métastases.The percentage rep resents the incidence of detectable CT-pr serum level in each group of patients. Patients ABCnEF<10*1930904861.180<102075041,6901,77535,700<101434339732,25111,600<1001001811.1001.91012,000<1001234515,6003.40068,000<100 10* " Serum levels measured before and 2 and 5 min after pentagastrin injection. ""8,, 10 m-IRMA CT-pr KC01-CT08 m-IRMA CT CT07-CT08 I (B) (A) Sfami 10» * In 1*31 ¡! IO4 .A. (D) 10s IO4 IO3 Fig. 7. CT-pr serum levels in patients with lung tumors. •.10 cases; •,O, +. 1 case; O. patient with detectable CT level (value indicated in parentheses); +, patient with liver métastases.The percentage represents the incidence of detect able CT-pr serum level in each group of patients. 10s IOIO" (H) 10- 5 10 15 20 25 Elution 5 10 15 20 25 fractions Fig. 5. Pattern of CT and CT-pr immunoreactivities in serum samples from a patient with MCT, as revealed by specific assays for either CT or CT-pr after a sequence of affinity chromatographies. [, beginning of elution with 2.5% acetic acid. A lo H refer to the fractions shown in Fig. 2. types, namely, squamous cell carcinoma, large cell carcinoma, adenocarcinoma, and undifferentiated carcinoma. Results pre sented in Fig. 7 show that the percentage of those patients with an elevated CT-pr level ranged from 17.5% (squamous cell carcinoma) to 53% (large cell carcinoma) and that only 2% of patients with non-SCLC lung tumors had simultaneous eleva tion of the serum CT level. We also determined CT-pr serum levels in patients with various malignancies (Fig. 8). While CTpr was undetectable in differentiated carcinoma of the thyroid, the percentage of positivity in the sera of other patients varied 6848 Downloaded from cancerres.aacrjournals.org on October 26, 2014. © 1989 American Association for Cancer Research. SERUM CALCITONIN PRECURSOR MEASUREMENT AND MALIGNANT DISEASE Fig. 8. CT-pr scrum levels in patients with non-MCT ma lignant tumors. • 10 cases; •.O, 1 case; O. patient with detectable CT level (value indicated in parentheses). The per centage represents the incidence of detectable CT-pr scrum level in each group of patients. widely, ranging from 7% (breast tumor) to 62% (hepatocellular carcinoma) depending on the cell type and stage of the tumor. Moreover, it was noteworthy that only 2 of these 298 patients had detectable levels of CT. Finally, we tested samples of patients with MCT and non-MCT malignant tumors at multiple dilutions to determine whether reactivity in the m-IRMA KC01-CT08 was comparable (parallel) to that of standard. Fig. 9 shows that the dose-response curve of the standard (PTN47) was parallel to those displayed by various serum samples of patients with MCT and non-MCT tumors taken as represent ative. Measurement of CI' and CT-pr Gene Transcript in a Hepato cellular Carcinoma. To confirm the production of CT-pr by a non-MCT tumor, we studied the presence of CT-pr at the tumor level. We prepared tumor extracts from both a normal liver tissue and a liver carcinoma of a patient displaying a serum CT-pr level of 11,000 pg/ml. Using the m-IRMA KC01-CT08, we did not find any significant binding in the extract of normal liver tissue, while we found a level of 28.2 pg/mg of tissue in the extract of liver carcinoma. Moreover, we performed hybrid ization experiments on the same tumor to detect CT-related RNAs. As a positive control in this experiment, we spotted an equivalent amount (25 ¿ig) of total RNA isolated from a human MCT. As a negative control, we spotted 25 /ug of yeast RNA. We saw more hybridization of the radiolabeled CT probe to tumor than to normal liver RNAs (data not shown). 10s p 10' 10' 10* - io-3 io-1 io-1 i Serum dilutions Fig. 9. Dose-response curves of standards (50.000 pg/ml) (•)and sera of patients with medullary carcinoma of the thyroid (x). leukemia (*). large cell lung carcinoma (A), small cell lung carcinoma (•).hepatocellular carcinoma (A). and neuroblastoma (*) at multiple dilutions. Serial Studies in Patients with Small Cell Lung Carcinoma. As patients with SCLC were those with the highest incidence of detectable CT-pr levels among patients with lung cancers (60%), we performed serial studies of CT-pr levels to evaluate the usefulness of CT precursors for the follow-up of these tumors. Simultaneously, we determined the serum levels of both CT and NSE, since the latter parameter has been described as a useful marker for follow-up of patients with SCLC (21). These studies were carried out in 3 patients during the course of their disease. In one patient, CT-pr levels remained consist ently elevated, whereas both CT and NSE levels were undetectable during the clinical disease-free period. In fact, a slight decrease in CT-pr levels during the first 2 mo of treatment was followed by a regular increase in levels up until evidence of liver métastasesby echographic examination. Similar results were observed in the other two patients, in whom elevation of CTpr levels preceded clinical or echographic evidence of recurring disease by 3 to 10 mo. However, in one of these patients, CT levels remained undetectable, while NSE and CT-pr levels increased simultaneously; in the other patient, both NSE and CT remained undetectable, while CT-pr levels increased. DISCUSSION Polyclonal antibodies directed to peptide hormones such as somatostatin, vasoactive intestinal polypeptide, and calcitonin appear to recognize heterogeneous immunoreactive molecules produced by various malignant tumors (22). For example, 59% of tumor tissues collected at random contain immunoreactive CT ( 18). Among these immunoreactive molecules are molecular species of higher molecular weight than those of native hor mones. Previous studies suggested that such species might well be precursor forms of these hormones (9). Hitherto, radioimmunoassays based on polyclonal antibodies were not capable of determining the precise nature of these forms. Recently, it was demonstrated that m-IRMAs enable the distinction of closely related gene products as well as the direct measurement of hormone precursors (12, 23). In fact, m-IRMA CT07-CT08 is an assay specific for mature CT, since antibody CT07 has the unique property of recognizing the 26-32 portion bearing a carboxamide function on its C-terminal amino acid residue as found on mature CT. The specificity of this assay was confirmed by the absence of reactivity of both CT-pr and KC in the mIRMA CT07-CT08. In contrast, the construction of m-IRMA KC01-CT08 based on one mAb directed to KC and the other one to CT enables the specific recognition of CT-pr. The specificity of this assay was confirmed by the lack of reactivity of both CT and KC in the m-IRMA KC01-CT08. We used such 6849 Downloaded from cancerres.aacrjournals.org on October 26, 2014. © 1989 American Association for Cancer Research. SERUM CALCITONIN PRECURSOR MEASUREMENT AND MALIGNANT DISEASE m-IRMAs to measure CT precursors separately from mature CT. Our results confirm the heterogeneity of immunoreactive CT in serum and demonstrate the presence of circulating CT precursors in patients with either benign or malignant diseases. Elevated values of up to 50,000 pg/ml were detected in 40% and 70rr of patients with benign liver disease and advanced gether, these latter results are consistent with the production and the excretion of CT precursors detected in serum by the tumor of this patient. Other studies suggested that many cancers are associated with ectopie hormone elaboration, but only a portion elaborate biologically active hormones and produce clinically recogniz renal failure, respectively, while all except one healthy subject able syndromes (27). Our data demonstrate that, at least in the had an undetectable serum CT-pr level (<100 pg/ml). The case of the calcitonin hormone, many cancers are associated presence of CT-pr in the blood of patients with renal failure with ectopie hormone precursor elaboration. The absence of might be due to delayed clearance of CT precursors; such a both a detectable circulating CT level and a clinically recogniz mechanism had already been suggested by the finding of mature able ectopie CT syndrome might be due to the lack of enzymes CT in patients with advanced renal failure (15, 24). In contrast, capable of metabolizing CT precursors to bioactive CT. Alter the isolated presence of CT-pr in 40% of patients with benign natively, the absence of complete processing of CT precursors liver diseases such as hepatitis or cirrhosis, although CT is in cancer patients might result from structural differences be consistently undetectable in such patients, might be due to an tween CT-pr present in their blood and precursors elaborated ectopie production of precursors by hepatic cells as well as to a in normal thyroid C-cells; modification in the biochemical defect in the hepatic metabolism of these molecules. structure would prevent the proteolytic cleavage of precursor Based on sera from patients with MCT, we established that molecules. Such alterations have been shown in the small cell neoplastic C-cells released incompletely processed CT precur lung carcinoma cell line DMS53, which biosynthesizes CT sors in addition to mature CT. It was noteworthy that, in those precursors. In these cells, glycosylation of CT precursors is not patients, CT-pr levels were correlated with CT levels, with the a feature of posttranslational processing, and the lack of gly former value being approximately 11 times higher than the cosylation seen in DMS53 cells might explain why apparently identical intracellular and extracellular precursors are detecta latter. This observation does not favor the hypothesis of alter ations in specific mechanisms of CT precursor processing in ble and perhaps why mature forms are only rarely present (9). Among the cancer patients with detectable CT-pr levels in those neoplastic C-cells. Such alterations would lead to marked variations in the ratios of precursors to mature hormones, as their blood, the highest incidence (62%) was found in patients observed in other tumor tissues containing prohormones and with hepatocellular carcinomas. Interestingly, 80% of patients hormones, for example, gastrinomas producing progastrin and with hepatocellular carcinomas, as well as 10 to 40% of patients with benign liver diseases, produced «-fetoprotein (28). Thus, gastrin (25). Moreover, it was striking that, after pentagastrin injection, the peak/basal ratio of CT-pr was lower than that of patients with either benign or malignant liver diseases have the capacity to produce molecules as different as oncofetal protein CT, suggesting that the reserve of precursor forms was smaller than that of mature CT in secretory granules of neoplastic C- or hormone precursors. A variety of substances, including sev eral peptide hormones such as calcitonin, were found to be cells. Previous studies performed by gel filtration separation followed by m-IRMA KC01-CT08 on sera of patients with present in serum from patients with lung cancers, particularly SCLC (29, 30). Biosynthesis of high-molecular-weight calci MCT demonstrated the presence of immunoreactive products with molecular weights of 14,000 and 8,000. Molecular weights tonin by human small cell carcinoma cells in tissue cultures was of these products were consistent with the hypothesis that the reported by several groups (9, 31, 32). We tested patients with KC01-CT08 combination binds to procalcitonin and to inter lung cancers of different histopathological types for the pres ence of CT and/or CT-pr in their blood and found the highest mediates of biosynthesis which include the CT sequence linked incidence (60%) of detectable CT-pr levels in patients with to the KC sequence (12). SCLC. Surprisingly, 53% of patients with large cell carcinomas In contrast to MCT tumors which produce both precursors also displayed an elevation in CT-pr serum levels. It is note and mature hormones, non-MCT tumors, in their large major worthy that these latter lung tumors do not belong to the ity, appeared to secrete CT precursors in the absence of mature CT. It was noteworthy that the dose-response curves observed neuroendocrine system as do SCLC, MCT, and other tumors with multiple dilutions of serum of patients with MCT and derived from Amine Precursor Uptake and Decarboxylation cells, a significant percentage of which produce CT-pr. How non-MCT tumors were parallel to that of standards. This observation strongly suggests that m-IRMA KC01-CT08 de ever, the histological typing of lung cancers has often been found to be difficult, and the histology of such cancers as well tects molecular species in sera comparable to the standard. as their capacity to produce CT-pr might change after therapy, Rises in serum CT-pr level in non-MCT patients were consist mutation, or new oncogene expression (33). Alternatively, the ent with previously reported ectopie secretion of immunoreac production of CT-pr by numerous lung tumors of various types tive CT (18, 26). This production of immunoreactive hormone might be explained by the existence of a common stem cell for was found with radioimmunoassays which measured any species reacting with a given polyclonal anti-CT antibody. Thus, it is all cell types; the existence of such stem cells has previously likely that a large number of tumors previously described as been suggested (34), and tumors deriving from these cells might producing "immunoreactive CT" do, in fact, produce CT pre have the common capacity to produce CT-pr. Since 60% of patients with SCLC had detectable CT-pr levels, cursors. In order to confirm the production of CT-pr by nonwe performed serial measurements of CT precursors to evaluate MCT tumors, we performed experiments with a liver carcinoma the usefulness of CT-pr determination for the follow-up of collected from a patient displaying a detectable serum CT-pr level in the absence of a detectable CT level. The finding of patients with SCLC. Our preliminary data showed that CT-pr immunoreactivity on the tumor extract with the m-IRMA levels paralleled changes in the clinical status and tumor burden KC01-CT08 strongly suggests that the material measured in of these patients. Interestingly, the rise in CT-pr levels preceded serum of this patient was produced by its tumor. Moreover, we evidence of tumor recurrence by several months. None of several found CT-related RNAs in this tumor. This observation shows biological markers previously proposed as indicators of either the extent of disease or the clinical response to cytotoxic therapy that the tumor was transcribing the calcitonin gene. All to 6850 Downloaded from cancerres.aacrjournals.org on October 26, 2014. © 1989 American Association for Cancer Research. SERUM CALCITONIN PRECURSOR MEASUREMENT appeared sensitive or specific enough for general use in the management of patients with SCLC. In light of our data, the measurement of CT-pr might be useful for the follow-up of these tumors. In conclusion, our study demonstrates that numerous patients with malignant tumors and, in particular, patients with tumors deriving from neuroendocrine cells had detectable levels of CT precursors. Finally, the development of m-IRMAs for different hormone precursors is important for measuring molecules that might be useful as tumor-associated markers and for under standing the processing of these precursors in cancer cells. ACKNOWLEDGMENTS We thank Dr. Jean-Michel Bidart and Dr. Alain Puisieux for their continuous help and valuable advice. We are grateful to Dr. Moukhtar for his thoughtful criticism of the data. We are indebted to Monique Benoit, Jean-Louis Bobot, Lionel Fougeat, and Yannick Smith for skillful technical assistance. We wish to thank Dr. Bidet for his help in the collection of sera from patients with lung tumors. REFERENCES 1. Austin. L. A., and Heath. H. III. Calcitonin: physiology and pathophysiology. N. Engl. J. Med.. 304: 269-278. 1981. 2. Le Moullec. J. M.. Jullienne. A., Chenais. J.. Lasmoles, F.. Guliana, J. M., Milhaud. G.. and Moukhtar, M. S. The complete sequence of human preprocalcitonin. FEBS Lett., 167: 93-97. 1984. 3. Bennett, H. P. J. Peptide hormone biosynthesis—recent developments. Re cent Results Cancer Res., 99: 34-45, 1985. 4. Douglass. J., Civelli. O., and Herbert. E. Polyprotein gene expression: generation of diversity of neuroendocrine peptides. Annu. Rev. Biochem., 53: 665-715. 1984. 5. Bradbury, A. F., Finnic, M. D. A., and Smyth, D. G. Mechanism of Cterminal amide formation by pituitary enzymes. Nature (Lond.). 298: 686688. 1982. 6. Deftos, L. J., Ross, B. A., Bronzert, D.. and Pathermore. J. G. Immunochemical heterogeneity 409-412. 1975. " of calcitonin in plasma. J. Clin. Endocrinol. Metab., 40: 7. Goodman. R. H.. Jacobs. J. W., and Habener. J. F. Cell-free translation of m-RNA coding for a precursor of human calcitonin. Biochem. Biophys. Res. Commun.. 91: 932-938. 1979. 8. Desplan. C.. Benicourt. C., Jullienne, A.. Segond. N.. Calmettes. C., Moukh tar, M. S., and Milhaud, G. Cell free translation of m-RNA coding for human and murine calcitonin. FEBS Lett.. 117: 89-92, 1980. 9. Cale. C. C.. Pettengill, O. S., and Sorenson, G. D. Biosynthesis of procalcitonin in small cell carcinoma of the lung. Cancer Res.. 46: 812-818. 1986. 10. Habener. J. F.. Singer, F. R.. Deftos. L. J.. and Potts, J. T.. Jr. Immunologieal stability of calcitonin in plasma. Endocrinology. 90: 952-958, 1972. 11. Motte, P.. Alberici, G.. Ait-abdellah. M., and Bellet. D. Monoclonal antibod ies distinguish synthetic peptides that differ in one chemical group. J. Immunol.. «A:3332-3338, 1987. 12. Ghillani. P.. Motte. P., Bohuon. C., and Bellet, D. Monoclonal antipeptide antibodies as tools to dissect closely related gene products: a model using peptides encoded by the calcitonin gene. J. Immunol.. 141:3156-3163, 1988. 13. Fraker, P. J.. and Speck, J. C.. Jr. Protein and cell membrane iodination using a sparingly soluble chloroamide. 1.3.4,6-tetrachloro-3a,6a-diphenyl glycuronyl. Biochem. Biophys. Res. Commun.. SO: 849-860. 1978. 14. Chomczynski. P.. and Sacchi, N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem.. 162: 156-159. 1987. AND MALIGNANT DISEASE 15. Motte. P.. Vauzelle. P., Gardet. P.. Ghillani. P.. Caillou. B.. Parmentier, C.. Bohoun. C., and Bellet. D. Construction and clinical validation of a sensitive and specific assay for serum mature calcitonin using monoclonal anti-peptide antibodies. Clin. Chim. Acta. 174: 35-54. 1988. 16. Pearse, A. G. E. The diffuse neuroendocrine system and the APUD concept: related "endocrine" peptides in brain, intestine, pituitary, placenta, and MUui.m cutaneous glands. Med. Biol.. 55: 115-125. 1977. 17. Sidhu, G. S. The endodermal origin of digestive and respiratory tract APUD cells. Am. J. Pathol.. 96: 5-20. 1979. 18. Abe. K., Adachi. I.. Miyakawa. S., Tanaka, M., Yamaguchi. K.. Tanaka. N., Kameya. T.. and Shimosato. V. Production of calcitonin, adrenocorticotropin hormone, and Ã-Ã--melanocyte-stimulatinghormone in tumors derived from amine precursor uptake and decarboxylation cells. Cancer Res.. 37: 41904194. 1977. 19. Heitz. P. U.. Klöppel.G.. Polak. J. M.. and Staub. J. J. Ectopie hormone production by endocrine tumors: localization of hormones at the cellular level by immunochemistry. Cancer (Phila.), 48: 2029-2037. 1981. 20. Gadzar. A. F.. Carney. D. N.. Becker. K. L.. Liang, V.. Go, W., Marangos. P. J., Moody, T. W., Wolfsen. A. R.. and Zweig. M. H. Expression of peptide and other markers in lung cancer cell lines. Recent Results Cancer Res.. 99: 167-174. 1985. 21. Carney. D. N.. Marangos. P. J.. Ihde. D. C.. Bunn. P. A., Jr.. Cohen. M. H., Minna, J. D., and Gadzar. A. F. Serum neuron-specific enolase: a marker for disease extent and response to therapy of small-cell lung cancer. Lancet, /: 583-585. 1982. 22. Sano, T., Saito, H.. Inaba, H.. Hizawa. K., Saito, S., Yamanoi, A., Mizumuma. Y.. Matsumura. M.. Yuasa. M., and Hiraishi. K. Immunoreactive somatostatin and vasoactive intestinal polypeptide in adrenal pheochromocytoma: an immunochemical and ultrastructural study. Cancer (Phila.). 52: 282-289. 1983. 23. Crosby, S. R., Stewart, M. F.. Ratcliffe, J. G., and While, A. Direct meas urement of the precursors of adrenocorticotropin in human plasma by twosite immunoradiometric assay. J. Clin. Endocrinol. Metab.. 67: 1272-1277, 1988. 24. Fauchald, P., Gautvik, V. T.. and Gautvik, K. M. Immunoreactive parathy roid hormone and calcitonin in plasma and ultrafiltrate before and after haemodialysis. Scand. J. Clin. Lab. Invest., 45: 229-235. 1985. 25. Del Valle, J., Sugano, K., and Y'amada, T. Progastrin and its glycine-extended 26. 27. 28. 29. 30. 31. 32. 33. 34. posttranslational processing intermediates in human gastrointestinal tissues. Gastroenterology. 92: 1908-1912. 1987. Krauss, S., Macy, S.. and Ichiki. A. T. A study of immunoreactive calcitonin (CT), adrenocorticotropic hormone (ACTH), and carcinoembryonic antigen (CEA) in lung cancer and other malignancies. Cancer (Phila.). 47: 24852492, 1981. Odell. W. D.. and Wolfsen. A. R. Hormones from tumors: are they ubiqui tous? Am. J. Med.. 68: 317-318. 1980. Bellet, D. H., Wands. J. R., Isselbacher, K. J., and Bohuon. C. Serum alphafetoprotein levels in human disease: a new perspective using a highly specific monoclonal radioimmunoassay. Proc. Nati. Acad. Sci. USA, 81:3869-3873, 1984. Yiakoumakis. E.. Proukakis. C.. Raptis. K.. Korres. A.. Dondas. N.. and Samaras, V. Calcitonin concentrations in lung cancer and non malignant pulmonary diseases. Oncology. 44: 145-149. 1987. Havemann. K., Luster. W.. Gropp, C., and Holle. R. Peptide hormone production associated with small cell lung carcinoma. Recent Results Cancer Res., 97: 65-76, 1985. Bertagna, X. Y., Nocholson, W. E.. Pettengill. O. S., Sorenson, G. D.. Mount. C. D., and Orth. D. N. Ectopie production of high molecular weight calcitonin and corticotropin by human small cell carcinoma cells in tissue culture: evidence for separate precursors. J. Clin. Endocrinol. Metab.. 47: 1390-1393, 1978. Zajac, J. D., Martin, T. J., Hudson, P.. Niall, H.. and Jacobs, J. W. Biosynthesis of calcitonin by human lung cancer cells. Endocrinology. 116: 749-755. 1985. Mabry, M., Nakagawa, T., Nelkin. B. D.. Me Dowell, E.. Gesell, M., Eggleston. J. C.. Casero. R. A., Jr.. and Baylin, S. B. v-Ha-ras oncogene insertion: a model for tumor progression of human small cell lung carcinoma. Proc. Nati. Acad. Sci. USA, 85: 6523-6527, 1988. Carney. D. N. The biology of lung cancer. Bull Cancer. 74: 495-500, 1987. 6851 Downloaded from cancerres.aacrjournals.org on October 26, 2014. © 1989 American Association for Cancer Research.