Main

Neuroblastoma is a childhood tumor of the sympathetic nervous system with an extraordinary clinical and biologic heterogeneity. The most favorable subset of this embryonal tumor may differentiate to ganglioneuroma or regress due to apoptosis after no or minimal therapy, whereas most metastatic neuroblastomas show progression and poor clinical outcome despite intensive multimodal therapy. Neuropeptides, a class of regulatory peptides produced by neural cells and acting as neurotransmiters, may have clinical and biologic significance in neuroblastoma and ganglioneuroma tumors(1). Neuropeptides may cause specific symptoms as well as regulate cellular growth and differentiation.

SOM is a cyclic neuropeptide with two different biologic active forms (SOM-14 and SOM-28) derived from a 92-amino acid precursor, presomatostatin. SOM is extensively distributed in the human body including the central and peripheral nervous system, the gastrointestinal tract, and various exocrine and endocrine glands(2–4). SOM has been shown to have growth inhibitory effects on malignant neuroendocrine cells in vitro, and long-acting analogues have been developed and used for treatment of neuroendocrine tumors in vivo(5,6).

Recent studies have shown that neuroblastoma cells may express SOM receptors with high-affinity binding, allowing biologic activities in vitro(7–9). These studies showed that octreotide may inhibit growth and induce apoptosis in neuroblastoma cells that express high-affinity receptors.

SOM receptors have also been identified in neuroblastoma tumor tissue, predominantly in samples from tumors of localized clinical stages without N-myc amplification(7,10). It was also shown that children with tumors expressing these receptors showed more favorable outcome(10). Recently, techniques for the detection of tumor cells expressing SOM receptors in vivo have been established(11). Preliminary results stated that SRS is positive at diagnosis or relapse in a majority of children with neuroblastoma(12). Our recent results show that positive SRS is associated with favorable tumor biology and favorable outcome(13). Further, it seems that SOM receptor expression may be under autocrine control and upregulated by SOM in vivo(14,15).

SH-SY5Y express receptors for IGF-I and IGF-II. Neuroblastoma tumors have been reported to proliferate in response to IGF-I(16). In acromegaly caused by pituitary tumors, octreotide often normalizes the elevated IGF-I levels(17), and serum IGF-I has been suggested as an intermediate marker for efficacy of SOM analogue treatment of pediatric tumors(18). We, therefore, were interested to find out if IGF-I could be used as an efficacy marker in octreotide treatment of neuroblastoma.

We recently reported that SOM is highly expressed in benign ganglioneuromas and in differentiated neuroblastomas(14,19). Both ganglioneuromas and neuroblastomas contain biologically active SOM molecular forms (SOM-14 and SOM-28)(20). In addition, SOM expression is associated with favorable prognosis in children >1 y of age with advanced neuroblastoma (International Neuroblastoma Staging System, INSS 3 and 4)(14).

In the present study, we have investigated the effect of the long-acting SOM analogue octreotide on the growth of a neuroblastoma cell line (SH-SY5Y) after inoculation in nude rats. SH-SY5Y xenografts expressed SOM receptors in vivo and responded to octreotide treatment with reduced tumor growth.

METHODS

Neuroblastoma cells. The adrenergic neuroblastoma cell line SH-SY5Y(21) was kindly provided by Dr. June Biedler, The Memorial Sloan-Kettering Cancer Center, New York. The cells were grown at 37°C in Eagle's minimum essential medium (Labsystems, Stockholm, Sweden) supplemented with 10% FCS (Sigma Chemical Co., St. Louis, MO), 1 mM L-glutamine, 100 IU/mL penicillin, and 50 µg/mL streptomycin in a humidified 5% CO2 atmosphere. The medium was changed twice a week and confluent cultures were subcultivated after 10 min of treatment with 0.25% trypsin and 0.02% EDTA. Cells were cultured on Nunclon Delta dishes (culture area 56.7 cm2, A/S Nunc, Roskilde, Denmark). To produce a large number of tumor cells, a confluent dish was seeded on a Nunclon triple flask with a culture area of 500 cm2. The cells were harvested when confluent. They were found by culture and by DNA staining to be free of mycoplasma.

For subcutaneous injections, a single cell suspension was prepared by treatment of the culture with 0.25% trypsin and 0.02% EDTA for 10 min, then suspended and centrifuged at 300 × g for 10 min. Before injection, the pellet was suspended in culture medium. Viable (i.e. trypan blue-excluding) cells were counted in a hemocytometer to achieve a final concentration of 100 × 106 cells/mL. The suspension was placed on ice.

Nude rats. Thirty-four nude rats (Rowett rnu/rnu), 20 males and 14 females, were used for xenografting at the age of 8-10 wk. The experiments were approved by the regional ethics committee for animal research. No animal was excluded from the study by the exclusion criteria applied, namely, loss of weight at two consecutive measurements and the development of an open wound over the tumor.

Xenografting. The animal was anaesthetized with 2% halothane (ISC Chemicals Ltd., Avonmouth, UK) supplemented with 50% N2O in oxygen. At injection, a 23-gauge cannula was used and care was taken to deposit the suspension subcutaneously without piercing the muscle fascia, and not to lose cells by leakage from the injection site. A small delineated wheal appeared at the injection site. Twenty million cells suspended in 0.2 mL of medium were injected in each hind leg.

Quantification of tumor growth. When tumor take was evident on palpation and/or was visible, the animal was anaesthetized with 2% halothane (ISC Chemicals Ltd.) supplemented with 50% N2O in oxygen. The tumor length (along the tumor long axis) and width (perpendicular to the long axis) were measured with a caliper by the same investigator throughout the experiments. It was not possible to make accurate measurements without anesthesia. Measurements were made every other day at 1000-1200 h. Tumor volume was calculated by length × width squared × 0.44(22). The true tumor weight was recorded at autopsy. Tumor volume index was calculated using the measured volume divided by the volume measured at tumor take at start of treatment.

Treatment with octreotide, retinoic acid, VIP, and untreated control group. When a tumor in an animal had reached a volume of 0.3 mL (designated d 0), the animal was randomized to one of four groups. Only those tumors that had reached a volume of 0.3 mL at the start of the treatment were followed up and evaluated for response. Six animals (eight tumors) received octreotide, 10 µg every 12 h s.c. adjacent to the tumor. Three animals (six tumors) received subcutaneous VIP (kindly provided by Professor Viktor Mutt at the Karolinska Institute), 40 µg daily, and four animals (seven tumors) were treated orally with 4 mg 13-cis-retinoic acid (which was a kind gift from Dr. Folke Bernadotte at Roche AG) once daily. Because of the different treatment strategies, the control group consisting of nine animals (14 tumors) did not receive any placebo treatment. Despite randomization at start of treatment, tumors in the VIP group had larger volumes than those treated with retinoic acid or octreotide. However, VIP-treated tumors did not differ from the control group at start of therapy.

IGF-I in serum. Serum from four rats from the octreotide-treated group and four rats from the control group were collected (intravenously) before treatment started, in the middle of the treatment period, and at the end of the treatment. Each time a sample was taken, the rats were anaesthetized as described previously. The samples were analyzed for concentrations of IGF-I in extracted serum by RIA by use of an antiserum raised in rabbits(23).

SRS. The animal was anesthetized with 2% halothane (ISC Chemicals Ltd.) supplemented with 50% N2O in oxygen and an injection with 0.2 mL of pentobarbital intraperitoneally. An average of 20 MBq 111In-pentreotide ([111In-DTPA-D-Phe1] octreotide, Octreoscan) was injected in the tail vein of the rat. Planar images were obtained after the injection at 4 and 24 h(24).

Statistics. Statistical analysis was performed using Mann-Whitney U test (2-sided probability) for two independent samples and t test when data were not nongaussian distributed. Kruskal-Wallis test with multiple comparisons was used for more than two groups. Friedman test was used for comparison of paired data in more than two groups.

RESULTS

SOM receptors in neuroblastoma xenografts. SOM receptor expression in vivo was investigated with 111In-pentetreotide in rats from all groups. All investigated rats showed labeling of tumor tissue, indicating high-affinity binding.

Treatment effects on tumor volume. Tumor volumes were measured every second day, and octreotide-treated neuroblastoma xenografts showed a reduced increase in tumor volume during treatment compared with untreated controls (Fig. 1). Tumor volumes of octreotide-treated rats were reduced after 10 and 12 d (mean 3.56 mL, range 1.38-6.80 mL and 4.24 mL, 1.31-6.59 mL, respectively) compared with untreated control tumors (6.48 mL, 2.32-14.48 mL and 8.01 mL, 4.50-11.56 mL, respectively, p < 0.001 and p < 0.005, respectively, Mann-Whitney). Retinoic acid-treated tumors were significantly smaller in terms of volume at the end of the treatment, whereas VIP-treated tumors did not differ from the untreated tumors.

Figure 1
figure 1

Neuroblastoma SH-SY5Y xenograft tumor volume in nude rats measured according to "Methods." Mean volumes at tumor take (d 0) and 2-12 d from start of treatment in two different groups of rats: octreotide-treated rats (○) and untreated control rats (▵). Tumor volumes for octreotide-treated rats were significantly smaller than the untreated tumors at 10 d (p < 0.001).

Full size image

Treatment effects on tumor weight. Octreotide therapy reduced the weight of tumors after 10-12 d of therapy (n = 8, median weight 2.90 g, range 1.67-5.57 g) compared with untreated control tumors (n = 14, median weight 7.54 g, range 1.65-10.82 g, p = 0.005) (Fig. 2). Tumors from rats treated with retinoic acid had significantly lower weight (n = 7, 4.30 g, 2.74-6.70) than tumors from untreated rats (p < 0.05, Mann-Whitney). Octreotide-treated tumors tended to have lower weight after treatment than retinoic acid-treated tumors (p = 0.09, Mann-Whitney). Tumors treated with VIP (n = 6, 7.43 g, 3.08-11.56) did not differ significantly from untreated controls.

Figure 2
figure 2

Neuroblastoma SH-SY5Y xenograft tumor weight at sacrifice 10-12 d from tumor take and start of treatment in two different groups of rats. CTRL indicates untreated control rats, and SOM indicates octreotide (SOM)-treated rats. Octreotide-treated rats had smaller tumors than untreated controls (p < 0.005).

Full size image

Serum IGF-I as a marker of octreotide therapy. Serum IGF-I decreased significantly (p < 0.001, Friedman test) over time in both the octreotide-treated rats and the untreated control rats (Fig. 3). There was no significant difference between the two groups.

Figure 3
figure 3

Concentrations of IGF-I in serum measured in rats before start of treatment (PRE) and during treatment after 7 and 12 d, respectively (POST). Both rats receiving octreotide (○ solid line) and untreated controls (▵, dashed line) showed significantly decreasing concentrations over time (p < 0.001). There was no difference between these two groups.

Full size image

DISCUSSION

SOM is known to inhibit neuroblastoma growth and to induce apoptosis in vitro. In the present study, we present data supporting the role of SOM in neuroblastoma therapy in vivo. We detected the presence of SOM receptors in human SH-SY5Y neuroblastoma xenografts grown in nude rats and a significant decrease in tumor growth from treatment with the long-acting SOM analogue octreotide.

The main focus of our investigation was to evaluate whether the treatment with a SOM analogue (octreotide) would affect the growth rate of neuroblastoma cells in vivo. Also, we wanted to compare SOM therapy with other agents with known or putative antineuroblastoma activity. Retinoic acid has been reported to induce differentiation of neuroblastoma cells under experimental conditions in vitro and in vivo(25–28).

In our present study, 13-cis-retinoic acid-treated tumors showed a significant reduction in weight after 10-12 d of therapy compared with untreated controls. There was a tendency indicating that octreotide reduced tumor growth more than retinoic acid in the dosages we used.

VIP has been shown to induce differentiation and to inhibit neuroblastoma growth in vitro(29). High concentrations of VIP have been detected in differentiated ganglioneuromas(19). Furthermore, elevated VIP in tumor tissue is associated with absence of N-myc amplification and a favorable outcome(19). Our present study aimed to investigate a putative effect of VIP in vivo in comparison with octreotide and 13-cis-retinoic acid. The limited number of tumors treated with VIP prevented any reliable interpretation of our data in this regard. In additional studies, we will investigate alternative VIP dosages and administration routes.

We were interested to see if measurement of serum IGF-I could be useful as a marker of octreotide treatment efficacy. However, the IGF-I concentrations decreased almost simultaneously over time in both the treated and the untreated group, so, unfortunately, IGF-I seems to be an inappropriate marker in this study.

Several investigators have shown that the presence of high-affinity SOM receptors visualized with scintigraphy is essential for a therapeutic effect of SOM analogues in tumor patients(30). Hence, our neuroblastoma model is a useful tool to investigate SOM effects, because high-affinity receptors are expressed in tumor tissue and the SOM peptide is present in very low amounts(14). Despite previous experiments indicating that the SH-SY5Y cell line only binds SOM with low affinity in vitro(8), we conclude that the therapeutic response in our present study (Figs. 1 and 2) is due to specific octreotide binding to high-affinity receptors in vivo.

Previous results in adult patients with neuroendocrine tumors indicate that SOM may have an autocrine up-regulating effect on specific SOM receptors in vivo(15). This may be in agreement with our recent preliminary findings in the neuroblastoma xenografts showing increased 111In-pentetreotide uptake in octreotide-treated tumors(14). However, the present study did not allow analyses to confirm or reject this hypothesis. We will further investigate the possible autocrine regulation of SOM receptors during octreotide treatment. Interestingly, O'Dorisio et al.(7) have reported receptor down-regulation during neuroblastoma tumor progression. Recent results showing absence of SOM receptors in most unfavorable neuroblastomas (with deletions of chromosome 1p and di/tetraploid DNA content) at diagnosis(13) may indicate that only a few children will be the focus of future SOM therapy. However, octreotide or other specific modalities may help to up-regulate essential receptors and enhance possibilities for SOM therapy of neuroblastoma. There may also be an effect of octreotide treatment despite negative Octreoscan imaging, as reported in a case of a carcinoid tumor(31).

From the present study, we conclude that SOM analogues may be active agents for in vivo therapy of neuroblastoma. Further studies are warranted to establish the role of SOM analogues in the treatment of children with neuroblastoma.