Expression of the somatostatin receptor subtype-2 gene predicts response of human pancreatic cancer to octreotide

Expression of the somatostatin receptor subtype-2 gene predicts response of human pancreatic cancer to octreotide William E. Fisher, MD, Peter Muscarella, MD, Thomas M. O'Dorisio, MD, M. Sue O'Dorisio, MD, PhD, Julian A. Kim, MD, Trisha A. Doran, BS, Carol L. Sabourin, Phi), and W'dliam J. Schirmer, MD, Columbus, Ohio

Background. Somatostatin inhibits proliferation of many solid tumors. The current study examines whether inhibition of the growth of pancreatic cancer by the somatostatin analog, octreotide, requires tumor expression of somatostatin receptors. Methods. We studied five human pancreatic cancer cell lines, Capan-1, Capan-2, CA V, MIA PaCa-2, and Panc-1. Solid tumors were established in nude mice (n = 20~cell line) by flank injection of tumor cells. Subcutaneous octreotide (500 I~g/kg/day) was administered by osmotic pumps to 10 of the animals in each group, and the other 10 received control infusions of saline solution. On day 36, the tumors were excised and weighed. Plasma levels of the putative trophic peptides cholecystokinin, epidermal growth factor (EGF), insulin-like growth factor-i (IGF-I), and insulin were assessed by radioimmunoassay. Each o~fthe fizve cell lines was assayed for the presence o~fcell sur~ace somatostatin receptors by using whole cell competitive binding assays with 125I-somatostatin. Expression of the somatostatin receptor subtype-2 (SSR2) gene was determined with reverse transcriptase-polymerase chain reactions. Southern blot hybridization was used to assess the presence of the SSR2 gene. Results. Octreotide inhibited tumor growth in the MIA PaCa-2 group (512 +- 75 mg control versus 285 + 71 mg treated; p < 0.05) but had no significant effect on tumor weight in the other four cell lines. Plasma levels of cholecystokinin, epidermal growth factor, insulin-like growth factor-l, and insulin were not altered by chronic octreotide infusion. Cell surface somatostatin receptors and SSR2 gene expression were detected only in the MIA PaCa-2 tumors. The gene for the SSR2 receptor was found in all five tumor lines. Conclusions. Octreotide-mediated inhibition of pancreatic cancer growth is dependent on expression of somatostatin receptors. The expression of somatostatin receptors should be considered in the design and interpretation of clinical trials with somatostatin analogs for treatment of pancreatic cancer. (Surgery 1 996; 120:234-4 I.) From the Departments of Surgery, Medicine, and Pediatrics, The Ohio State University, Columbus, Ohio

SOMATOSTATINHAS BEENCHARACTERIZEDas the universal off-switch because it inhibits the release of growth horm o n e a n d virtually all gastrointestinal hormones. Recent studies with experimental t u m o r models of pancreatic cancer have shown an antiproliferative effect of somatostatin both in vitro and in vivo. 1 These studies suggest that somatostatin a n d its analogs might serve as Supported by grantsfrom The AmericanCancer Society,The Bremer Foundation, National Cancer Institute, National Research Service AwardCA-09338,Divisionof CancerPreventionand Control,and Surgical Research Inc. Presented at the Fifty-seventhAnnual Meetingof the Societyof UniversitySurgeons, Washington,D.C., Feb. 8-10, 1996. Reprintrequests:WilliamJ. Schirmer,MD, The Ohio State University, N724 Doan Hall, 410 W. Tenth Ave.,Columbus,OH 43210. Copyright 9 1996 by Mosby-YearBook, Inc. 0039-6060/96/$5.00 + 11/6/73520 234

SURGERY

relatively nontoxic agents in adjuvant therapy of pancreatic cancer. T h e m e c h a n i s m by which somatostatin inhibits proliferation of solid tumors has not b e e n established. Inhibition may be mediated directly via specific, high-affinity somatostatin receptors o n the t u m o r cell surface. Alternatively, somatostatin may slow t u m o r growth indirectly by inhibiting the secretion of trophic h o r m o n e s including several growth-promoting gastrointestinal peptides. Additional theories include somatostatin-mediated inhibition of t u m o r angiogenesis a n d selective inhibition of splanchnic a n d t u m o r blood flow. This study examines whether inhibition of pancreatic cancer growth by the somatostatin analog, octreotide, is d e p e n d e n t on t u m o r expression of somatostatin receptors. T h e influence of octreotide o n the growth of five h u m a n pancreatic cancer cell lines is characterized in

Surgery Volume 120, Number 2

nude mice. The effect of octreotide is correlated with tumor expression of somatostatin receptors and with changes in plasma levels of several trophic peptides. We hypothesize that octreofide suppresses pancreatic cancer growth directly via somatostafin receptors on the tum o r cell surface.

METHODS Cell lines. The h u m a n pancreatic adenocarcinoma cell lines Capan-1, Capan-2, MIA PaCa-2, and Panc-1 were purchased from the American Type Culture Collection (Rockville, Md.), and CAV was a gift from Dr. Courtney Townsend (Galveston, Texas). Capan-1 was cultured in Roswell Park Memorial Institute-1640 medium with L-glutamine and 20% fetal bovine serum. Capan-2 was cultured in McCoy's 5A medium with 10% fetal bovine serum. The MIA PaCa-2 and Panc-1 cell lines were grown in Dulbecco's modified Eagle medium with 10% fetal bovine serum. Horse serum (2.5%) was added to the MIA PaCa-2 culture medium. The CAV cells were grown in a 1:1 mixture of Roswell Park Memorial Institute-1640 medium and Dulbecco's modified Eagle medium with 10% fetal bovine serum. All cell culture materials were purchased from Gibco BRL, Gaithersburg, Maryland. All incubations were at 37 ~ C, with 95% humidity and 5% CO,2 tension. In vivo study. One hundred nude mice ( N u / n u ; Harlan-Sprague-Dawley, Indianapolis, Ind.) were randomly assigned to one of five tumor groups (n-- 20/ group). The human pancreatic cancer cell lines in the log phase of growth were treated with trypsin and 5 x 106 cells were injected subcutaneously in the flank of each animal. On the following day the animals in each group were randomly assigned to a treatment (n = 10) or control (n = 10) group. Alzet osmotic pumps (model 2002; Alza Corp., Palo Alto, Calif.) were loaded with octreotide (1 m g / m l ; Sandoz, East Hanover, N.J.) that was diluted with normal saline solution to deliver a dose of 500 p g / k g body weight/day. The pumps were implanted subcutaneously on the flank opposite the tumor by using a sterile technique. Animals in the control groups received pumps loaded with normal saline solution. Every 2 weeks the empty pumps were removed and replaced with fresh pumps by using the same subcutaneous pocket. This was repeated for 36 consecutive days of continuous subcutaneous octreotide infusion. We encountered no problems with infection using this technique. The tumors were excised from the anesthetized animals (50 m g / k g intraperitoneal pentobarbital; Abbott, Chicago, Ill.). The tumors were immediately weighed and snap frozen in liquid nitrogen. The animals were promptly exsanguinated. The blood was collected in ethylenediamine tetraacetic acid (EDTA) tubes and kept on ice until the plasma portion could be frozen. The plasma concentrations of cholecystokinin,

Fisher et aL

235

(EGF), insulin-like growth factor-1 (IGF-1), insulin, and octreotide were measured by double-antibody radioimmunoassay. 2 Within each tumor group, differences between the treated and control groups were determined by the Mann-Whitney Unonparametric test. Differences were considered significant at the p < 0.05 level. In vitro receptor assays. The presence of receptors for somatostatin was determined with classic competitive binding assays on whole cell preparations. Cells harvested from tissue culture were seeded into 6-well plates (60,000 cells/well). The cells were allowed to grow to confluence (48 to 72 hours). The confluent monolayers were washed three times with serum-free medium, and 400 Ill of Dulbecco's modified Eagle medium was added to each well. 125I-somatostatin-14 (50 lal, 1000 CPM/pl; Peninsula Labs, Inc., Belmont, Calif.) was added to every well. Cold somatostatin-14 (50 pl; Peninsula Labs, Inc.) in increasing concentrations from 0 to 10-4 m o l / L was added to the wells in triplicate. The plates were placed in the incubator (37 ~ C, 5% CO2, 95% humidity) for 1 hour on a slowly rocking table. The cell monolayers were then washed three times with serum-free medium and centrifuged. The radioactivity in the cell pellet was determined with a g a m m a radiation counter (Beckman 5500; Beckman Instruments Inc., Fullerton, Calif.). Each experiment was done three times for a total of nine observations per concentration. The radioactivity was expressed as percentage of total binding. The raw data were analyzed with the LIGAND curve-fitting software package (Elsevier-Biosofi, Cambridge, U.K.) and Scatchard analysis to calculate the dissociation constants (Ka) of the receptors, z~ In vitro proliferation assays. The growth response of each of the five h u m a n cell lines to somatostatin-14 was determined by incorporation of tritiated thymidine. The cells were harvested from tissue culture, and 5000 cells were seeded into each well of a 96-well plate. After 24 hours the cells had adhered to the bottom of the wells. The cells were washed with serum-free medium, and 90 pl minimum growth medium (a 1:4 dilution of the standard medium) was added to each well. Ten microliters of serum-free medium was added to each of the 16 wells in the first two columns, which served as the untreated controls. Somatostatin-14 was added to the 16 wells in the treated columns in final concentrations of 10 -7 m o l / L , 10-9 m o l / L , and 10-11 m o l / L . At 24-hour intervals the cells were washed and fresh somatostau'-fi was added. After 72 hours of growth 1 pCi tritiated thymidine (New England Nuclear, Boston, Mass.) in 100 pl medium was added to each well. After an additional 6 hours of incubation the cells were frozen. O n a subsequent day the cells were harvested from the wells onto glass fiber filter paper (M.A. Bioproducts, Whittaker Corp., Walkersville, Md.) by using the PHD Cell Harvester (Cambridge Technology, Inc., Watertown, Mass.).

236

Fisher et al.

Surgery August 1996

Table I. Effect of chronic octreotide infusion on h u m a n pancreatic cancer implants in nude mice

Control Treated p Value

Capan-1

Capan-2

CAV

MIA PaCa-2

Panc-1

76 - 10 70 -+ 12 NS

286 +- 44 218 -4-49 NS

1859 -+ 442 1991 -+ 502 NS

512 - 75 285 + 71 0.043

476 -+ 128 375 - 128 NS

Final tumor weight (rag) 36 days after tumor cell implantation is expressed as mean ~- SEM in control (n = 10) and octreotide treated (n = 10) groups. NS, Not significant, *p < 0.05 by Mann-Whitney U test.

Table II. Effect of chronic subcutaneous octreotide infusion on plasma levels of potential growth factors

Control Treated p Value

Cholecystokinin (pg/ml)

EGF (pg/ml)

IGF-1 (ng/ml)

Insulin (ttU/ml)

Octreotide (pg/ml)

140 --- 27 153 -4-27 NS

23 -+ 6 32 +- 6 NS

276 +- 19 266 - 19 NS

35 -+ 3 39 + 3 NS

40 - 4 3029 -+ 358 <0.001

Plasma levels of each peptide on day 36 of octreofide infusion are expressed as mean _+ SEM in control (n = 10) and octreotide treated (n = 10) groups. NS, Not significant. *p < 0.05 by Mann-Whitney U test.

The disks of filter paper were placed in vials with 5 ml Bio-Safe II liquid scintillation cocktail (Research Products International Corp., Mount Prospect, Ill.) and allowed to stand overnight. Scintillation counting was performed on the following day with the Beckman LS 1801 beta spectrometer (Beckman Instruments Inc.). The results were expressed as percentage change from the untreated control. Overall statistically significant differences a m o n g the treatment groups were determined with the Kruskal-Wallis nonparametric test. If differences existed, posteriori pairwise comparisons were made with the Bonferroni test. Differences were considered significant at the p < 0.05 level. U s e o f reverse transcriptase-polymerase chain reaction to detect somatostatin receptor gene expression. The snap-frozen tumor specimens were ground to a fine powder in liquid nitrogen. Total RNA was isolated from the powdered tumors by using TRIzol solution (GibcoBRL) according to the manufacturer's protocol. PhaseLock Gels (5 Prime-3 Prime, Inc., Boulder, Colo.) were used to simplify separation of the phenol-chloroform phase from the aqueous phase. All RNA samples were analyzed for integrity of 18S and 28S ribosomal RNA (rRNA) by ethidium bromide staining of 0.5 lag RNA resolved by electrophoresis on 1.2% agarose formaldehyde gels. Reverse transcription was performed for 60 minutes at 37 ~ C in a volume of 20 pl containing 2 lag RNA, 50 units Moloney murine leukemia virus reverse transcriptase in 10 m m o l / L Tris HC1 p H 8.3, 50 m m o l / L KC1, 5 m m o l / L MgCI2, 1 unit/lal ribonuclease inhibitor (RNAsin), 1 m m o l / L each of deoxyadenosine triphosphate, deoxyguanosine triphosphate, deoxycytidine triphosphate, and deoxythymidine triphosphate and 100 pmoles of r a n d o m hexamers. The samples were

preincubated for 10 minutes at 25 ~ C and heated to 99 ~ C for 5 minutes to terminate the reverse transcription reaction. For cDNA amplification by the polymerase chain reaction (PCR), 10 !11of the first strand synthesis reaction was added to the amplification mixture. Final amplification mixtures contained 10 m m o i / L Tris HC1 pH 8.3, 50 m m o l / L KC1, 2.5 m m o l / L MgC12 0.2 m m o l / L each of deoxyadenosine triphosphate, deoxyguanosine triphosphate, deoxycytidine triphosphate, and deoxythymidine triphosphate, 0.25 p m o i / L each of sense and antisense primer and 1.25 units Taq DNA polymerase (Gibco-BRL) in a final volume of 50 pl. The reaction mixtures were heated to 95 ~ C for 30 seconds. Thirty amplification cycles were performed at 95 ~ C for 15 seconds, at 60 ~ C for 30 seconds, and at 72 ~ C for 30 seconds, followed by an incubation for 7 minutes at 72 ~ C. By use of the published somatostatin receptor subtype-II (SSR2) sequence, PCR primers specific for h u m a n SSR2 (sense, 5'-TGCTGGGTCTGCCTTI'CTI'G-3' and antisense, 5'-GGTCTCCGTGGTCTCATFCA-3') were chosen and generated a 772 base pair PCR product. 4 Primers for the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which generated a 710 base pair PCR product, were used to control for the a m o u n t of RNA in each reaction. Water and reverse transcriptase minus reactions were run as negative controls. The amplification products were analyzed on a 1.8% agarose gel. U s e o f Southern blot to detect somatostatin receptor gene. A portion of the powdered tumor specimens was incubated overnight with digestion buffer (12 p l / m g tumor). The digestion buffer consisted of proteinase K (0.1 m g / m l ) , 100 m m o l / L NaC1, 10 m m o l / L Tris-C1 p H 8.0, 25 m m o l / L EDTA p H 8.0, and 0.5% so-

Surgery Volume 120, Number 2

Fisher et al.

237

100 - . - Capan-1

80

- , - Capan-2 % TB

-*- CAV

60

--- MIA --.- Panc-1

40 20 0

10 "13 10 "12 10 "11 10 "1~ 10 .9

10 "8

SOMATOSTATIN

10 .7 10 4

10 s

(M)

Fig. 1. Cell surface receptor studies. Somatostatin concentration in molarity is expressed on the X-axis. Radioactivity is expressed on the Y-axis as percentage of total binding (TB). Data presented as mean _+ SEM (n = 9 / d a t a point). Only the MIA PaCa-2 cell line possesses cell surface somatostafin receptors, Kd = 3 -+ 2 nmol/L.

110 - . - Capan-1 - , - Capan-2

% Control

-*- CAV 90

-.- MIA - . - Panc-1

70

i

0

10 "11

i

i

10 .9

SOMATOSTATIN

10-7

(M)

Fig. 2. Effect ofsomatostatin on proliferation of human pancreatic cancer cells in culture. Somatostatin-14 concentration is expressed in molarity on the X-axis. Effect of 72-hour incubation on incorporation of tritiated-thymidine is expressed on the Y-axis as percentage of untreated control (n = 16/group). *p < 0.05 by Kruskal-Wallis with posteriori comparisons by Bonferroni. Somatostatin inhibited growth of only the MIA PaCa-2 cell line.

d i u m dodecyl sulfate (SDS). G e n o m i c DNA was extracted from the digested t u m o r samples with a mixture of p h e n o l a n d chloroform and precipitated with 0.1 volume o f 7.5 m m o l / L a m m o n i u m acetate a n d 2 volumes o f 95% ethanol. Twenty micrograms o f the DNA was digested overnight at 37 ~ C with EcoR1 restriction enzyme (50 units) in React III (1• buffer (GibcoBRL). T h e digested DNA was subjected to electrophoresis t h r o u g h a 75 ml 1% agarose gel. T h e gel was washed twice with 0.25 N HC1 for 7 minutes, rinsed in distilled water, a n d washed for 10 minutes in 0.4 N N a O H . T h e DNA was transferred onto a Biodyne-A nylon membrane (Gibco-BRL) by using a standard overnight blotting set-up with 0.4 N N a O H as the buffer. T h e

m e m b r a n e was washed twice for 10 minutes in 2x stand a r d saline citrate (SSC)-0.2 m o l / L Tris p H 7.5 a n d twice for 5 minutes in 0.1 x SSC-0.1% SDS. T h e membrane was prehybridized in Hybrisol-1 (Oncor, Gaithersburg, Md.) containing 1.5 m g d e n a t u r e d salmon sperm DNA (Gibco-BRL) a n d hybridized overnight at 42 ~ C in Hybrisol-1 with a r a n d o m - p r i m e d , 32p-labeled PCR p r o d u c t for SSR2 g e n e r a t e d by reverse transcriptase-polymerase chain reaction (RT-PCR) as described in the previous section. T h e SSR2 cDNA could be used as a g e n o m i c DNA p r o b e because the SSR2 g e n e lacks introns. T h e identity o f the SSR2 PCR p r o d u c t was confirmed by restriction f r a g m e n t analysis. T h e m e m brane was washed twice at r o o m t e m p e r a t u r e for 5 min-

238

Fisher et al.

tO

tO tO

Surgery August 1996

a.

SSR2 (772 bp)

GAP (710 bp)

Fig. 3. RT-PCR to detect somatostatin receptor gene expression. RNA was isolated from frozen tumor specimens and transcribed to cDNA with reverse transcriptase. SSR2 gene was amplified by PCR. Housekeeping gene GAPDH was used to control for loading error. Only the M1A PaCa-2 cell line expresses mRNA for the SSR2 gene.

utes with 2• SSC-0.1% SDS. The membrane was exposed to Hyperfihn-MP (Amersham, Arlington Heights, Ill.) at -70 ~ C for 8 hours by using intensifying screens.

RESULTS Injection of the five h u m a n pancreatic cancer cell lines resulted in tumor formation in 100% of the nude mice. Treatment with octreotide significantly inhibited tumor growth in vivo in the MIA PaCa-2 group; however, no significant inhibition of tumor growth was noted in the other pancreatic cancer groups (Table I). Octreotide levels in the treatment group were 3029 + 358 p g / m l compared with 40 +-4 p g / m l in the control group (p < 0.05). Peripheral plasma levels of cholecystokinin, EGF, IGF-1, and insulin were not altered by octreotide treatment (Table II). Competitive binding assays revealed somatostatin receptors on the surface of the MIA PaCa-2 cells. The binding curves for l'25I-somatostatin on Capan-1, Capan-2, CAV, and Panc-1 were flat, indicating the absence of somatostatin receptors in these four cell lines (Fig. 1). In vitro proliferation of MIA PaCa-2 was blunted significantly by somatostatin treatment. Somatostatin had no

effect in vitro on the somatostatin-receptor negative cell lines (Fig. 2). Expression of the SSR2 gene was shown by RT-PCR in the MIA PaCa-2 tumors but not in the other groups (Fig. 3). Southern blot hybridization showed the presence of the SSR2 gene in all five tumor types (Fig. 4). In summary, growth of only one of the five h u m a n pancreatic cancer lines, MIA PaCa-2, was inhibited both in vivo and in vitro by octreotide. Differences in growth rates could not be explained on the basis of suppression of circulating levels of the trophic pepfides cholecystokinin, EGF, IGF-1, or insulin. Expression of the SSR2 gene, as detected by RT-PCR, correlated with the presence of cell surface somatostatin receptors, as determined by whole cell competitive binding. Furthermore, SSR2 gene expression appeared to predict the inhibition of tumor growth by octreotide.

DISCUSSION The only curative treatment for pancreatic cancer is complete resection. Unfortunately, curative resection is rarely feasible and is associated with recurrence rates of nearly 90%. Current radiation and chemotherapy regimens offer a marginal survival advantage but negatively affect the quality of life of the patient. Somatostatin analogs are free of major side effects. If efficacy of these c o m p o u n d s can be established, they offer an attractive addition to traditional adjuvant therapy for pancreatic cancer. The antiproliferative action of somatostatin and its analogs on pancreatic cancer has been shown in recent in vitro and in vivo studies. Schallf' reported that the somatostatin analog RC 160 reduced the weight and volume of the tumor and prolonged host survival in nitrosamine-induced pancreatic carcinoma in Syrian golden hamsters. Upp et al. 6 demonstrated inhibition of two h u m a n pancreatic cancers, SKI and CAV, mainmined as nude mouse xenografts, by the administration of octreotide. O u r data confirmed a favorable response in selected cell lines, but the favorable response was not universal. For example, we did not observe growth inhibition of CAV by octreotide either in vitro or in vivo. The promising animal data spurred clinical trials of somatostatin analogs in the adjuvant treatment of pancreatic cancer. Unfortunately, results have been disappointing. Klijn et al. 7 treated 14 patients who had metastatic pancreatic cancer with three daily subcutaneous injections of octreotide (100 to 200 pg/injection) for an average of 7 weeks and observed no antitumor effect. Nineteen patients with advanced exocrine pancreatic carcinoma were given the somatostatin analog BIM 23014 by using a range of doses from 250 lag a day to 1 mg a day for 2 months. One patient had a partial response, 6 patients had stable disease, and 11 had progressive disease, s Huguier et al., ~ in a randomized pro-

Surgery Volume 120, Number 2

spective study o f 86 patients who were given a similar treatment regimen, d e m o n s t r a t e d no significant increase in m e d i a n survival rates according to life-table analysis. An explanation for the discrepancy between the animal and h u m a n trials has not b e e n elucidated. It is possible that a higher dose of somatostatin and its analogs should have been used in the h u m a n trials. Many o f the animal studies used doses at least an o r d e r o f m a g n i t u d e greater than those used in the clinical trials. O u r data suggest that expression o f the somatostatin rec e p t o r is critical for response to octreotide. We propose that the main reason for the failure o f adjuvant treatm e n t with somatostatin analogs is that most h u m a n pancreatic cancers lack receptors for somatostatin. Clinical studies thus far have n o t e x a m i n e d the tumors for receptors or a t t e m p t e d to correlate response with r e c e p t o r status. Although several gastrointestinal h o r m o n e s and o t h e r growth factors including insulin, cholecystokinin, secretin, gastrin, EGF, and IGF-1 have been shown to p r o m o t e growth o f the n o r m a l exocrine pancreas and exocrine pancreatic cancer, this study does not s u p p o r t the postulate that somatostatin works by inhibiting trophic peptides.l~13 T h e observation that high doses o f octreotide failed to suppress peripheral levels of many of these peptides is in keeping with clinical observations. Klijn et al. 7 measured insulin, IGF-1, a n d EGF levels in their clinical study of the use of octreotide for pancreatic cancer. Chronic treatment with a somatostatin analog had no effect on EGF levels. Although they observed early significant decreases in insulin and IGF-1 levels, the levels o f both growth factors had r e t u r n e d to p r e t r e a t m e n t values by 5 days and 4 weeks, respectively. This study does not support suppression of trophic peptides as an i m p o r t a n t mechanism by which somatostatin inhibits pancreatic cancer, but it does not eliminate the possibility that these h o r m o n e s may influence pancreatic t u m o r growth. A subcutaneous location o f the tumor, instead of in the pancreas or liver, may significantly influence the findings. T h e concentration of many o f these peptides is dramatically h i g h e r within the pancreatic milieu and portal system c o m p a r e d with the p e r i p h e r a l circulation. T h e current study does not address the possibility of intrapancreatic paracrine influences. This study suggests that pancreatic cancers with somatostatin receptors can be p r e d i c t e d to r e s p o n d favorably to octreotide treatment. In addition, the absence o f somatostatin receptors predicts failure of octreotide therapy. Using competitive b i n d i n g assays a n d RT-PCR, we e x a m i n e d an additional five h u m a n pancreatic a d e n o c a r c i n o m a cell lines, AsPC-1, BxPC-3, CFPAC-1, Hs766T, a n d SU.86.86. N o n e of the five additional cell lines had cell surface somatostatin receptors,

Fisher et al.

X--" =U

O4 |

0

0

239

"" |

i

0

a.

Fig. 4. Southern blot to detect somatostatin receptor gene. Genomic DNA was isolated from frozen tumor specimens, digested with EcoR1, subjected to electrophoresis, transferred to a nylon membrane, and hybridized with 32P-labeled cDNA probe for SSR2 gene. SSR2 gene is present in tumor DNA from all five human pancreatic cancers. expressed the SSR2 gene, or r e s p o n d e d to somatostatin in vitro (unpublished data). T h e absence o f somatostatin receptors on most pancreatic cancers is a likely explanation for the failure o f pancreatic cancer to r e s p o n d to adjuvant t r e a t m e n t with somatostatin analogs. T h e somatostatin r e c e p t o r is expressed by the n o r m a l h u m a n pancreas, whereas most o f the h u m a n pancreatic cancers we have studied do not have somatostatin receptors. 14 T h e mechanism does n o t a p p e a r to involve deletion o f the SSR2 gene because its presence was exhibited in all five cell lines. T h e molecular mechanism responsible for down-regulation o f SSR2 gene expression could occur at the transcriptional or translational level. Because expression o f the SSR2 gene appears to predict the response of pancreatic cancer to octreotide, strategies to increase SSR2 gene expression in tumors lacking receptors may prove beneficial. This study d e m o n s t r a t e d that expression o f the SSR2 gene correlated with a response to octreotide. Five somatostatin r e c e p t o r subtypes were cloned. T h e SSR2 subtype a p p e a r e d to be the most a b u n d a n t l y expressed subtype in n o r m a l h u m a n tissues a n d h u m a n tumors. 15 Although octreotide has b e e n f o u n d to b i n d with greatest affinity to the SSR2 receptor, it also binds to the SSR3 a n d SSR5 r e c e p t o r s ) 6' a7 We have not conclusively proved that SSR2 is the only somatostatin r e c e p t o r responsible for suppression o f t u m o r growth by octreotide. F u r t h e r studies are n e e d e d to assess the importance o f the additional r e c e p t o r subtypes.

240

Fisher et al.

We conclude that somatostatin-mediated inhibition o f p a n c r e a t i c c a n c e r g r o w t h is d e p e n d e n t o n e x p r e s s i o n o f s o m a t o s t a t i n r e c e p t o r s . Clinical trials o f s o m a t o s t a t i n a n a l o g s as a d j u v a n t t r e a t m e n t f o r p a n c r e a t i c c a n c e r should include an assessment of the somatostatinr e c e p t o r status o f t h e t u m o r s . W e h y p o t h e s i z e t h a t m o s t

Surgery August 1996 16. Raynor K, Murphy WA, Coy DH, et al. Cloned somatostatin receptors: identification of subtype-selectivepeptides and demonstration of high affinity binding of linear peptides. Mol Pharmacol 1993;43:838-44. 17. Raynor K, O'Carroll AM, Kong H, et al. Characterization of cloned somatostatin receptors SSTR4 and SSTR5. Mol Pharmacol 1993;44:385-92.

p a t i e n t s with p a n c r e a t i c c a n c e r will n o t b e n e f i t f r o m a d j u v a n t t r e a t m e n t with s o m a t o s t a t i n a n a l o g s b e c a u s e

DISCUSSION

o f t h e a b s e n c e o f s o m a t o s t a t i n r e c e p t o r s . A small s u b s e t

Dr. R. Daniel Beauchamp (Nashville, Tenn.). I think that your somatostatin results are very interesting. I do have a question about some of your data. You implied that there were no EGF receptors on the Panc-1 cells, the MIA cells, and the other cell lines that you assayed, and I am really concerned about that result. About 10 years ago Murray Korc showed that these pancreatic cancer cell lines overexpressed the EGF receptor, and we confirmed that a couple of years later. As a matter of fact, they have about 1 million EGF receptors or more per cell, and that is very easily shown by radioreceptor binding assays and looking at receptor protein by immunoprecipitation. So why were you unable to detect any EGF receptor binding on these cells? Dr. Fisher. We hypothesized that EGF receptors would be present, and in fact we chose the MIA cell line because it has been shown before by Schally (Proc Natl Acad Sci U S A 1991;88:1658-60) to have EGF receptors, and we were somewhat surprised by our results. From our data we conclude that there are no receptors for EGF on all five of our cell lines. O n e explanation for these unexpected results is that the cells stop expressing EGF receptors after numerous passages in cell cultttre. However, even if EGF receptors were present, treatment with octreotide failed to suppress plasma levels of EGF. It is unlikely tltat the mechanism of growth inhibition by octreotide is by suppressing EGF secretion. Dr. D.H. Teitelhaum (Ann Arbor, Mich.). Do you think one additional set of experiments whereby you again block the somatostatin receptor with an antibody to somatostatin receptor and then demonstrate a lack of somatostatin inhibition of your MIA cell line might be the last confirmation that that is truly the mechanism in which somatostatin acts on these cells? Second, how do you explain the difference between previous authors that have shown an inhibition of growth in the CAV cell line, like Dr. Townsend's group, and your study, which does not show this inhibition of cell growth? Dr. Fisher. With regard to your first question about neutralizing the somatostatin receptors with antibody, we have thought of that study. One other possibility would be to transfect the cells with a gene that would encode for antisense message for SSR2 and in that way prevent translation o f just the type-It receptor. We are actually approaching it from another angle. We have ongoing studies where we are transfecting the gene for SSR2 into the four receptor negative cell lines, and we hypothesize that by putting the receptor on the negative cell lines we will then make them responsive to octreotide. With regard to your second question about why the CAV cell line was not inhibited by somatostatin as it was in Dr. Townsend's laboratory, one explanation may be that the tumor they were using was being passed through n u d e mice as a subcutaneous implant and the tumor we received from their laboratory was an immortal cell line established from

o f p a t i e n t s with s o m a t o s t a t i n - r e c e p t o r positive t u m o r s , h o w e v e r , m i g h t r e s p o n d favorably. S o m a t o s t a t i n r e c e p t o r g e n e e x p r e s s i o n s h o u l d b e c o n s i d e r e d in t h e d e s i g n a n d i n t e r p r e t a t i o n o f clinical trials with s o m a t o s t a t i n a n a l o g s as a d j u v a n t t r e a t m e n t f o r p a n c r e a t i c c a n c e r . REFERENCES 1. Evers MB, Parekh D, Townsend CM, Thompson JC. Somatostatin and analogues in the treatment of cancer: a review.Ann Surg 1991;213(3):190-8. 2. Morgan CR, Lazarow A. Immunoassay of insulin: two antibody system--plasma insulin levelsof normal subdiabetic and diabetic rats. Diabetes 1963;12:115-26. 3. Munson PJ, Rodbard D. Ligand: a versatile computerized approach for characterization of ligand-binding systems.Anal Biochem 1980;107:220. 4. Yamada Y, Post SR, Wang K, Tager HS, Bell GI, Seino S. Cloning and flmctional characterization of a familyof hnman and mouse somatostatin receptors expressed in hrain, gastrointestinal tract, and kidney. Prnc Natl Acad Sci USA 1992;89:251-5. 5. SchallyAV. Oncological applications of somatostatin analogues. Cancer Res 1988;48:6977-85. 6. UppJR, Ol~m D, Pnston FJ, et al. Inhibition of growth of two truman pancreatic adenocarcinomas ill viw:, by somatostatin analog SMS 201-9995. AmJ 8nrg 1988;155:29-35. 7. Kl!jnJGM, Hoff AM, Th AS, et al. Treatment of patients witb metastatic pancreatic and gastrointestinal tnmours with tile somatostatin analogue Samlostatin: a phase 11 study including endocrine ettects. BrJ Cancer 1990;62:627-30. 8. Canobbio L, Boccardo F, Cannata D, Galloni P, Epis R. Treatment oI" advanced pancreatic carcinoma with the somatnstatin analogue BIM 23014. 1992;69:648-50. 9. Hugnier M, Samama G, TestartJ, et al. Treatment of adenocarcinoma of the pancreas with somatostatin and gnnadoliberin (lutenizing hormone-releasing bnrmone). Am J Surg 1992; 164:348-53. 10. Liehr RM, MelnykovychG, Solomon TE. Growth effects of regulatory peptides on human pancreatic cancer cell lines Panc-I and MIA PaCa-2. Gastroenterology 1990;98:1666-74. 11. Townsend CM, Franklin RB, Watson LC, GlassEJ, ThompsonJC. Stimulation of pancreatic cancer growth by caerulein and secretin. Surg Forum 1989;32:228-9. 12. Watson SA, Crosbee DM, Morris DL, et al. Therapeutic effect of the gastrin receptor antagonist, CR2093, on gastrointestinal tumor cell growth. BrJ Cancer 1992;65:879-83. 13. Smith JR, Kramer S, Bagheri S. Effects of a high-fat diet and L364,718 on growth of human pancreas cancer. Dig Dis Sci 1990;35:726-32. 14. Fekete M, Zalatnai A, Comaru-Schally AM, Schally AV. Membrane receptors for peptides in experimental and human pancreatic cancers. Pancreas 1989;4:521-8. 15. Woltering EA, O'Dorisio MS, O'Dorisio TM. The role of radiolabeled somatostatin analogs in the management of cancer patients. Principles and practice of oncology: npdate 1995;9:1-16.

Surgery Volume 120, Number 2

those tumors. In the process of establishing an immortal cell line, a less differentiated cell that does not have somatostatin receptors may have been selected. Dr. F. Charles Brunicardi (Houston, Texas). How can you be sure that you are dealing with SSR2. As you know there are five somatostatin receptors that have been cloned, and all five of them interact with somatostatin. How can you be sure that your probe is specific for human somatostatin subtype-II? Furthermore, message expression is only suggestive of the specific receptor subtype on the cell surface. Dr. Fisher. Somatostatin-14, which we used in the competitive binding assays, binds to all five of the somatostatin receptor subtypes, so we can say with certainty that the receptornegative cell lines do not have any of the five somatostatin receptors. We concluded carefully that somatostatin receptors are important in the response to octreotide, but we did not go so far as to say that it was just SSR2. We have proved that the type-II receptor gene is transcribed by the responding cell line,

Fisher et al.

241

but we have not proved that types I, III, IV and V are not there. We are sure that the primers for SSR2 are specific to the type-II receptor because they code for a sequence in the type II receptor gene that is not present in the other four receptor subtypes. Dr. Daniel T. Dempsey (Philadelphia, Pa.). Was there any survival benefit to somatostatin in the MIA mice? Did you follow any of these animals out and did they survive better? Dr. Fisher. We e n d e d the study before death of any o f the animals. We chose a subcutaneous location for the tumor because it is very rare for the tumors to metastasize from this location and it also gives the opportunity to measure tumor area and plot a growth curve as the study progresses. It might be interesting to implant the MIA tumors in the pancreas itself and study the effect of octreotide treatment on survival. Another somatostatin analog, RC-160, has been shown to prolong survival of Syrian hamsters with nitrosamine-induced pancreatic cancer.