Advertisement
Journal of Clinical Oncology  
Search for:
Limit by:
  Browse by Subject or Issue
Home Search or Browse JCO My JCO Subscriptions Customer Service Site Map

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a colleague
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Save to my personal folders
Right arrow Download to citation manager
Right arrowRights & Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhu, Z.
Right arrow Articles by Büchler, M. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhu, Z.
Right arrow Articles by Büchler, M. W.
Journal of Clinical Oncology, Vol 17, Issue 8 (August), 1999: 2419
© 1999 American Society for Clinical Oncology

Nerve Growth Factor Expression Correlates With Perineural Invasion and Pain in Human Pancreatic Cancer

Zhaowen Zhu, Helmut Friess, Fabio F. diMola, Arthur Zimmermann, Hans U. Graber, Murray Korc, Markus W. Büchler

From the Department of Visceral and Transplantation Surgery and Institute of Pathology, University of Bern, Inselspital, Switzerland, and Departments of Medicine, Biological Chemistry, and Pharmacology, University of California, Irvine, CA.

Address reprint requests to Helmut Friess, MD, Department of Visceral and Transplantation Surgery, University of Bern, Inselspital, CH-3010 Bern, Switzerland; email helmut.friess{at}insel.ch


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
PURPOSE: The reasons for the high frequency of perineural invasion and the presence of pain in pancreatic cancer are still not clear. Nerve growth factor (NGF) and its high-affinity receptor TrkA are involved in stimulating epithelial cancer cell growth and perineural invasion, as well as in pain generation in chronic benign disorders.

PATIENTS AND METHODS: NGF and TrkA were examined by Northern blot analysis, in situ hybridization, and immunohistochemistry in 27 normal and 37 pancreatic cancer tissue samples. The molecular findings were correlated with the degree of perineural invasion, pain, and histopathologic tumor characteristics.

RESULTS: Northern blot analysis indicated that NGF and TrkA mRNA levels were increased 2.7-fold and 5.6-fold, respectively (P < .05 and P < .05), in pancreatic cancer tissues compared with the normal pancreas tissue. As shown by in situ hybridization and immunohistochemistry, NGF was strongly present in the cytoplasm of pancreatic cancer cells. TrkA was intensely present in the perineurium of pancreatic nerves but not in the cancer cells. There was no difference in NGF and TrkA expression between early (stages I and II) and advanced (stage III) tumor stages and between well-/moderately differentiated (grades 1 and 2) and poorly differentiated (grade 3) tumors. However, tumors with high NGF/TrkA expression levels exhibited more frequent perineural invasion (P < .01). Furthermore, increased NGF/TrkA expression levels were associated with a higher degree of pain (P < .01).

CONCLUSION: Enhanced expression of the NGF/TrkA system may influence perineural invasion and may contribute to the pain syndrome in human pancreatic cancer.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
PANCREATIC CANCER IS presently the fourth or fifth leading cause of cancer death in Western countries, with a 5-year survival rate of around 10% to 25% after tumor resection.1 Perineural invasion extending to the extrapancreatic nerve plexus, the most common route of spread for pancreatic cancer cells, is a histopathologic characteristic in pancreatic cancer; it leads to retropancreatic tumor extension, precludes curative resection, and influences local recurrence after tumor resection.2-11 The affinity of pancreatic cancer cells to infiltration of neural structures is particularly striking, and incidences of between 84% and 100% underscore its importance.2-4

There are various hypotheses to explain perineural invasion in pancreatic cancer. Earlier, it was speculated that cancer cells simply extended their growth along perineural and intraneural lymph vessels.12 However, later studies on animals and humans did not confirm the presence of lymph vessels along pancreatic nerves. Therefore, it was suggested that cancer cells first extended along a plane of lowest resistance, which was thought to be the perineural space.13 More recent concepts proposed that a suitable microenvironment of growth advantage may exist in the perineural space, enhancing tumor cell proliferation and tumor cell spread.14 However, it seems that perineural invasion is not simply a consequence of abundant innervation of the pancreas and less resistance in the perineural space. The critical element may be that the cancer cells are attracted and their growth is stimulated by factors—such as transforming growth factor–alpha, epidermal growth factor receptor, and neural cell adhesion molecule—which are present especially in enlarged pancreatic nerves.14-16 The presence of these factors in pancreatic nerves strongly supports the hypothesis that the perineural space provides a suitable microenvironment for the growth and chemotaxis of tumor cells in pancreatic cancer and in other malignancies in which perineural invasion is observed.

Nerve growth factor (NGF) is a neurotrophic protein that affects the development and survival of a variety of neural cell types in both the peripheral nervous system and CNS.17 Experiments in cultured lung cancer cells, a prostate cancer cell line, and a human glioblastoma cell line have indicated stimulatory effects of NGF on tumor growth, tumor invasion, and the formation of metastases.18-20 Subsequently, it was speculated that NGF may influence the process of perineural invasion in human prostatic adenocarcinomas.21

TrkA is the high-affinity receptor for NGF and is an essential component in the mediation of NGF response.22 This NGF receptor is a protein tyrosine kinase of approximately 140 kd of molecular mass encoded by the TrkA proto-oncogene.23 The binding of NGF to TrkA induces receptor autophosphorylation and activation of intracellular signaling pathways, resulting in diverse biologic effects.23

In the present study, the expression and distribution of NGF and TrkA in human pancreatic cancer tissues at mRNA and protein levels were investigated, and the molecular findings were correlated with the presence of perineural invasion, pain, and histopathologic tumor characteristics.


   PATIENTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Patients and Tissue Collection
Pancreatic cancer tissues were obtained from 37 patients (17 women and 20 men) undergoing a partial duodenopancreatectomy (Whipple resection) for pancreatic cancer. The median age of the pancreatic cancer patients was 69 years (range, 49 to 83 years). According to the tumor-node-metastasis classification and histopathologic grading system of the International Union Against Cancer,24 there were nine stage I, seven stage II, 21 stage III, and zero stage IV tumors. Tumor grading showed nine well-differentiated tumors, 22 moderately differentiated tumors, and six poorly differentiated tumors (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Clinicopathologic Parameters and NGF and TrkA Expression in Pancreatic Cancer Tissue Samples
 

Normal human pancreatic tissue samples were obtained from 27 individuals (16 women and 11 men) who were free of pancreatic disease through an organ donor program in which there were no candidates for pancreatic transplantation. The median age of the organ donors was 45 years (range, 18 to 57 years). All normal tissue samples were obtained from the head of the organ donor's pancreas to ensure comparability with the tumor samples. Freshly removed tissue samples were immediately fixed in Bouin's or paraformaldehyde solution for 12 to 24 hours and embedded in paraffin for immunohistochemistry and in situ hybridization. Concomitantly, tissues for RNA extraction were snap frozen in liquid nitrogen in the operating room upon surgical removal and maintained at -80°C until use. The studies were approved by the human subject committee of the University of Bern, Bern, Switzerland.

Analysis of Perineural Invasion
The presence of perineural invasion was assessed in all pancreatic cancer specimens by two independent observers blinded to patient status; any differences were resolved by joint review and consultation with a third observer. For each cancer sample, about 10 tissue sections from different tumor locations were checked. Perineural invasion was defined as positive if the infiltration of cancer cells into the perineurium or neural fasciculus was detected at the leading point, as reported previously.6,25 The degree of perineural invasion was defined microscopically as follows: 0, no perineural invasion; 1, perineural invasion was difficult to find, with <= one occurrence per slide; 2, perineural invasion was easy to find, with two to four occurrences per slide; and 3, perineural invasion was easy to find, with more than four occurrences per slide or intraneural invasion (Table 1).

Assessment of Pain
Pain was assessed in all patients before surgery using a standardized questionnaire. The degree of pain was defined as follows: 0, no pain; 1, mild pain (abdominal discomfort or pain not requiring analgesics or not disabling); 2, moderate pain (pain controlled by nonnarcotic analgesics); and 3, severe pain (pain that required narcotic analgesics and was disabling) (Table 1).

Northern Blot Analysis
Total RNA was extracted using the single-step guanidinium isothiocyanate method,26 followed by electrophoresis under denaturing conditions in a 1.2% agarose/1.8 mol/L formaldehyde gel, as previously reported.27 The gels were stained with ethidium bromide for verification of RNA integrity and loading equivalency. The RNA was electrotransferred onto nylon membranes (Gene Screen; Du Pont, Boston, MA) and cross-linked by ultraviolet irradiation. The filters were then prehybridized, hybridized, and washed under highly stringent conditions. The blots were prehybridized overnight at 65°C in 50% formamide, 0.5% sodium dodecyl sulfate (SDS), 5x standard saline citrate (SSC), 5x Denhardt's solution (1x Denhardt's = 0.02% Ficoll, 0.02% polyvinylpyrrolidone, and 0.02% bovine serum albumin), 250µg/mL salmon sperm DNA, and 50 mmol/L sodium phosphate (pH 6.5). The blots were then hybridized for 18 hours at 65°C in the presence of 1 x 106 cpm/mL of the 32P-labeled antisense NGF and TrkA cRNA probes, washed twice at 65°C in 1x SSC and 0.5% SDS, and washed twice at 65°C in 0.1x SSC and 0.5% SDS.

To assess equivalent RNA loading, all blots were rehybridized with a mouse 32P-labeled 7S cDNA probe, which cross-hybridizes with human 7S RNA.27 The blots were prehybridized overnight at 42°C in a buffer containing 50% formamide, 1% SDS, 0.75 mol/L NaCl, 5 mmol/L EDTA, 5x Denhardt's solution, 100µg/mL salmon sperm DNA, 10% dextran sulfate, and 50 mmol/L sodium phosphate (pH 7.4). Hybridization followed under the same conditions, with exposure to 1 x 105 cpm/mL of the 32P-labeled 7S cDNA probe and then two washes at 50°C in 2x SSC and three washes at 55°C in 0.2x SSC and 2% SDS.

All blots were exposed at -80°C to Fuji x-ray film (Tokyo, Japan) with Kodak intensifying screens (Rochester, NY) for 1 to 10 days. The intensity of the radiographic bands was quantified by a computerized video system and the Image-pro-plus 3.0 software (Media Cybernetics, Silver Spring, MD). The ratios of the optical densities of the RNA levels (NGF:7S and TrkA:7S) were calculated for each sample.

In Situ Hybridization
In situ hybridization was performed as previously reported.28 Briefly, pancreatic tissue samples were fixed in paraformaldehyde and embedded in paraffin. The tissue sections (2 to 4µm) were deparaffinized, dehydrated, and incubated in 0.2 mol/L HCl for 20 minutes. The sections were treated with proteinase K (Boehringer Mannheim, Mannheim, Germany) at a concentration of 50µg/mL for 15 min at 37°C. After fixation with 4% paraformaldehyde in phosphate-buffered saline for 5 min, the samples were prehybridized at 50°C for at least 1 hour in 50% formamide (v/v), 4x SSC, 2x Denhardt's solution, and 250µg of RNA/mL. Hybridization was performed overnight at 50°C in 50% (v/v) formamide, 4x SSC, 2x Denhardt's solution, 500µg of RNA/mL, and 10% dextran sulfate (w/v). The final concentrations of the digoxigenin-labeled probes were approximately 0.5 ng/µL. After hybridization, the sections were washed and treated with RNase (Boehringer Mannheim). The samples were then incubated with an anti-digoxigenin antibody conjugated with alkaline phosphatase (Boehringer Mannheim; dilution 1/500). For color reaction, 5-bromo-4-chloro-3-indolyl phosphate and nitro-blue-tetrazolium (Sigma, Buchs, Switzerland) were used. For control experiments, the slides were incubated with RNase or with the corresponding sense probes. Pretreatment of the slides with RNase abolished the hybridization signal produced by the antisense probe. Furthermore, incubation with the sense probe failed to produce in situ hybridization signals.

cRNA and cDNA Probe Synthesis
A 305-base pair (bp) KspI/PstI fragment of the human NGF cDNA and a 401-bp SacI/KspI fragment of human TrkA cDNA were generated by reverse transcriptase polymerase chain reaction, and the polymerase chain reaction fragments were subcloned into the pGEM-T Easy vector (Promega Biotechnology, Madison, WI) carrying promoters for the DNA-dependent SP6 and T7 RNA polymerases. The authenticity of the subcloned DNA fragment was confirmed by sequencing, using the dye terminator method (ABI 373A; Perkin Elmer, Rotkreuz, Switzerland).

The 7S probe consisted of a 190-bp BamHI/BamHI fragment of mouse 7S cDNA, which was subcloned into the pGEM 7ZF(+) vector (Promega Biotechnology) and which cross-hybridizes with human 7S.

For Northern blot analysis, NGF and TrkA antisense cRNA probes were radiolabeled with [alpha-32P]-cytidine triphosphate (Du Pont), and the 7S cDNA probe was labeled with [alpha-32P]-deoxycytidine triphosphate (Du Pont).

NGF and TrkA cRNA probes were labeled with digoxigenin and used for in situ hybridization.28 After linearization, the DNA was transcribed using the Ribomax system (Promega Biotechnology). The transcription resulted in digoxigenin-labeled antisense riboprobes specific for the NGF and TrkA mRNA. The corresponding sense probes were prepared using an analog method.

Immunohistochemistry
Consecutive 3- to 5-µm paraffin-embedded tissue sections were subjected to immunostaining using the streptavidin peroxidase technique (Kirkegaard & Perry Laboratories, Inc, Gaithersburg, MD), as previously reported.29 After deparaffinization and dehydration, the tissue sections were submerged for 15 minutes in Tris-buffered saline (TBS) solution (10 mmol/L Tris-HCl and 0.85% NaCl; pH 7.4) containing 0.1% (vol/vol) Triton X-100 and then washed for 5 minutes in TBS buffer. Endogenous peroxidase activity was blocked by incubating the slides in methanol and in methanol/0.6% hydrogen peroxide, followed by one washing in methanol and two washings in TBS containing 0.1% bovine serum albumin. After treatment with hyaluronidase (1 mg/mL in 100 mmol/L sodium acetate and 0.85% NaCl), the sections were incubated for 30 minutes at 37°C with 10% normal goat serum and then incubated overnight. Incubation was carried out at 4°C, with the primary antibodies diluted in 10% normal goat serum as follows: polyclonal rabbit anti-NGF (Serotec Ltd, Oxford, United Kingdom) (1/600 dilution) raised against ultrapure 2.5 S NGF; polyclonal rabbit anti-TrkA immunoglobulin (Ig) (Santa Cruz Biotechnology, Santa Cruz, CA) raised against an epitope corresponding to amino acids 763 to 777 mapping adjacent to the carboxyl terminus of human Trk p140 (1/600 dilution). The antibody has no cross-reactivity with TrkB or TrkC. NGF and TrkA antibodies were pretested for immunostaining and immunoblotting to confirm their specificity.30,31 Bound antibody was detected with a biotinylated goat anti-rabbit IgG (secondary antibody) and a streptavidin-peroxidase complex (Kirkegaard & Perry Laboratories), followed by incubation with diaminobenzidine tetrahydrochloride (0.05%) as the substrate and counterstaining with Mayer's hematoxylin. To ensure specificity of the primary antibodies, consecutive sections were incubated either in the absence of the primary antibody or with a nonimmunized rabbit IgG antibody. In these cases, no immunostaining was detected. Histopathologic analysis was performed by two independent observers blinded to patient status; any differences were resolved by joint review and consultation with a third observer.

To evaluate the influence of NGF and TrkA on pain and perineural invasion, the immunohistochemical findings were scored in a semiquantitative fashion, as previously described.32 In the case of NGF, the intensity of immunostaining and the percentage of immunoreactive cancer cells were analyzed. In the case of TrkA, the intensity of immunostaining in the perineurium and the percentage of nerves with immunoreactivity in the perineurium were analyzed. The staining intensity was recorded as follows: 0, no immunostaining; 1, weak immunostaining; 2, moderate immunostaining; and 3, intense immunostaining. The immunohistochemical staining score for each sample was calculated as intensity x the percentage of positive cells.

Statistical Analysis
For statistical analysis, the Mann-Whitney test and Fisher's exact test were used. The relationships among the variables were assessed using linear regression analysis and calculation of the Spearman's correlation coefficient. In all cases, significance was defined as P < .05.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Northern Blot Analysis of NGF
Northern blot analysis in 27 normal pancreas samples and 37 ductal pancreatic cancer tissue samples was performed using 20µg of total RNA. In the normal pancreas, the 1.3-kb NGF transcript was weakly visible on the original autoradiographs in eight (30%) of 27 tissue samples, whereas in the other normal tissue samples, NGF mRNA expression was too low to be detectable. In comparison with the normal pancreas, 21 (57%) of 37 pancreatic cancer samples overexpressed the 1.3-kb NGF mRNA transcript. The expression levels in these samples showed a wide range. In the remaining pancreatic cancer samples, NGF mRNA signals were detectable only on the original autoradiographs. Densitometric analysis of the expression signals revealed a 2.7-fold increase (P < .05) in NGF mRNA levels in pancreatic cancer samples compared with the normal control samples when all cancerous tissue samples were included. When only cancer samples with increased NGF mRNA expression levels were statistically analyzed, there was a 4.8-fold increase (P < .01) (Fig 1).



View larger version (48K):
[in this window]
[in a new window]
  		Fig 1. Northern blot analysis of NGF and TrkA gene expression in normal and in pancreatic cancer tissue samples. Blots were rehybridized with a 32P-labeled 7S cDNA probe to verify equivalent RNA loading. NGF and TrkA were concomitantly overexpressed in pancreatic cancer tissues compared with normal control samples.

 

Northern Blot Analysis of TrkA
Low levels of TrkA mRNA expression were found in the normal pancreas. The 3.2-kb TrkA mRNA transcript was weakly detectable in 15 (56%) of 27 normal tissue samples. Comparative analysis of NGF and TrkA in the normal pancreas samples revealed that all samples which exhibited NGF mRNA expression also showed TrkA mRNA expression (Fig 1). There were seven normal samples in which there was TrkA mRNA expression but no NGF mRNA expression was detectable.

Enhanced TrkA mRNA expression was detected in 22 (59%) of 37 pancreatic cancer samples. The densitometric analysis of the expression signals revealed that when all cancer samples were analyzed together, there was a 5.6-fold increase (P < .05) in TrkA mRNA levels in the pancreatic cancer samples in comparison with the normal control samples. When only cancer samples with enhanced TrkA mRNA signals were analyzed, there was a 9.4-fold increase above normal (P < .01) (Fig 1). All but one of the cancer samples exhibited concomitant overexpression of NGF and TrkA mRNA.

In both the normal pancreas and pancreatic cancer, there was a positive correlation between NGF mRNA and TrkA mRNA levels (normal pancreas: r = .79, P < .01; pancreatic cancer: r = .95, P < .01).

In Situ Hybridization
To determine the exact site of NGF and TrkA mRNA expression, in situ hybridization was performed in normal and cancer samples in which mRNA signals were detectable by Northern blot analysis (Table 2). In all experiments, tissue sections of normal and cancerous samples were processed simultaneously to ensure comparability of the results. In addition, sense NGF and TrkA cRNA probes were used on consecutive tissue sections to ensure specificity of the in situ hybridization signals.


View this table:
[in this window]
[in a new window]
  		Table 2. Localization of NGF and TrkA mRNA and Protein in the Normal Pancreas and in Pancreatic Cancer Tissue Samples
 

NGF mRNA signals were moderately present in the cytoplasm of most ductal cells and weakly present in the cytoplasm of acinar cells in normal pancreatic parenchyma (Fig 2A). However, nerve fibers did not show any NGF mRNA signals (Table 2). In contrast, in pancreatic cancer tissues, NGF mRNA signals were strongly present in the cytoplasm of pancreatic cancer cells (Fig 2C). As in the normal pancreas, no NGF mRNA signals were present in the nerves (Fig 2E). The remaining normal pancreas surrounding the cancer mass showed a NGF staining pattern comparable to that of the normal control samples.



View larger version (104K):
[in this window]
[in a new window]
  		Fig 2. In situ hybridization of NGF in (A) normal pancreas and in (C, E, G) pancreatic cancer tissue samples and in situ hybridization of TrkA in (B) normal pancreas and in (D, F, H) pancreatic cancer. The arrows in panel F point to the perineurium. Original magnification x200; n, nerve; a, artery; v, vein.

 

TrkA mRNA expression was weakly present in the perineurium of most nerves in the normal pancreas (data not shown). Normal ductal and acinar cells were also devoid of any TrkA mRNA signals (Fig 2B). In contrast, in pancreatic cancer tissue samples, TrkA mRNA signals were moderately detectable in the perineurium of nerves (Fig 2F), whereas pancreatic cancer cells were devoid of any TrkA mRNA signals (Fig 2D).

NGF and TrkA mRNA expression were also present in intrapancreatic ganglia, with stronger signal intensity in pancreatic cancer samples than in the normal pancreas (data not shown). The endothelial cells of arteries and veins and the muscle layer of arteries in both the normal pancreas tissues (data not shown) and pancreatic cancer tissues (Fig 2G and 2H) exhibited intense NGF and TrkA mRNA signals, with no difference in signal intensity and frequency between the normal and the pancreatic cancer samples.

Immunohistochemistry
In the normal pancreas, moderate NGF immunostaining was present in the cytoplasm of most ductal cells and weak NGF immunostaining was present in the cytoplasm of acinar cells (Fig 3A). In pancreatic cancer tissues, NGF immunoreactivity was intensely present in the cytoplasm of pancreatic cancer cells (Fig 3E), whereas nerve fibers in both the normal pancreas and pancreatic cancer exhibited moderate NGF immunostaining (Fig 3C and 3G) (Table 2).



View larger version (99K):
[in this window]
[in a new window]
  		Fig 3. Immunostaining of NGF in (A, C) the normal pancreas and in (E, G) pancreatic cancer and immunostaining of TrkA in (B, D) normal and (F, H) pancreatic cancer tissue samples. Panels C and D, panels E and F, and panels G and H are consecutive tissue sections. Original magnification of A, B, E, F, G, and H, x200, and C and D, x100; n, nerve; a, artery; v, vein. Arrows indicate the perineurium.

 

TrkA immunoreactivity was rarely found in the normal pancreas; it was found mainly in moderate intensity in the perineurium of nerves (Fig 3D). No TrkA immunostaining was present in ductal and acinar cells (Fig 3B). Pancreatic cancer cells were also devoid of any TrkA immunostaining (Fig 3F). However, in pancreatic cancer tissues, strong TrkA immunoreactivity was present in the perineurium of pancreatic nerves (Fig 3H). The perineurium of nerves infiltrated by pancreatic cancer cells was often severely damaged, and therefore the intensity of the immunohistochemical signal in these regions could not be evaluated conclusively (Fig 3F).

NGF and TrkA immunostaining was also present in the intrapancreatic ganglial cells, where the signals were stronger in pancreatic cancer samples than in normal control samples (data not shown). The endothelial and muscle cells of arteries and the endothelial cells of veins in both the normal pancreas tissue (Fig 3C and 3D) and pancreatic cancer tissue (Fig 3G and 3H) exhibited moderate NGF and intense TrkA immunostaining. The signal intensities and frequencies of normal and pancreatic cancer samples were similar (Table 2).

Relationship of NGF and TrkA Expression With Clinicopathologic Parameters
The NGF and TrkA mRNA expression and protein levels obtained by Northern blot analysis and by immunohistochemistry in cancer samples were analyzed in relation to clinicopathologic parameters of the patients (Table 1). There was no difference in NGF and in TrkA mRNA levels between early (stages I and II) and advanced (stage III) tumor stages (P > .05 and P > .05, respectively) and between well-/moderately (grades 1 and 2) and poorly (grade 3) differentiated pancreatic cancer samples (P > .05 and P > .05, respectively). However, NGF and TrkA expression levels were significantly higher (P < .01 and P < .01, respectively) in pancreatic cancer samples with perineural invasion than in cancer samples without perineural invasion. Furthermore, patients with pain had significantly higher NGF (P < .01) and TrkA mRNA (P < .01) expression levels in their tumors than did patients without pain. The same results were obtained when these calculations were done with NGF and TrkA immunostaining.

In analyses of the degree of perineural invasion and the degree of pain with regard to the NGF and TrkA mRNA levels, there was a positive relationship between the degree of perineural invasion and NGF mRNA levels (r = .66, P < .01) and between the degree of perineural invasion and TrkA levels (r = .58, P < .01). Furthermore, there was a positive relationship between pain and NGF mRNA levels (r = .63, P < .01) and between pain and TrkA mRNA levels (r = .64, P < .01).

When the degree of perineural invasion and the degree of pain were analyzed with regard to the NGF and TrkA protein levels (immunohistochemical staining score), there was a positive relationship between the degree of perineural invasion and the NGF immunohistochemical staining score (r = .63, P < .01) and between the degree of perineural invasion and the TrkA immunohistochemical staining score (r = .62, P < .01). Furthermore, there was a positive relationship between pain and the NGF immunohistochemical staining score (r = .67, P < .01) and between pain and the TrkA immunohistochemical staining score (r = .64, P < .01).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In the present study, NGF and TrkA mRNA levels were markedly increased when pancreatic cancer samples were compared with normal pancreas samples, using Northern blot analysis. Further study using in situ hybridization and immunostaining demonstrated that NGF was highly expressed in pancreatic cancer cells, whereas TrkA was not present in pancreatic cancer cells and was mainly present in the perineurium of pancreatic nerves. Therefore, it seems that NGF released from the cancer cells has no autocrine and paracrine effects on cancer cells itself but might interact with TrkA, which is present in the perineurium, and that there may exist some attraction between pancreatic cancer cells and pancreatic nerves.

The frequent infiltration of pancreatic cancer cells in nerves has been noted for a long time,2 and recent data indicate that in pancreatic cancer patients, the perineural invasion is significantly associated with the prognosis.5,6 However, the mechanisms that contribute to invasion of pancreatic nerves by cancer cells and to the spread of cancer cells along the nerves are poorly understood. Newer concepts, based on recent molecular findings, proposed a direct interaction of pancreatic cancer cells with nerves, and growth factors,14 cell adhesion molecules,15,16 and gastrointestinal hormones have been suggested to promote the migration of pancreatic cancer cells along nerves.8 The statistical analysis in our study showed that NGF and TrkA mRNA expression levels were significantly higher in tumors with perineural invasion than in tumors without perineural invasion, and that in tumors with high NGF and TrkA expression, perineural invasion was more frequent. Therefore, it may be suggested that in pancreatic cancer there is close interaction between the cancer cells and the neural system, and that the NGF/TrkA pathway may have an important role in the process of perineural invasion and the spread of cancer cells along nerves. This hypothesis is supported by a previous study in which it was shown that the NGF receptor is present in cells of the perineurium, and that axonal regeneration and nerve growth are stimulated by NGF binding to the NGF receptor in the perineurium.33 NGF can act as a positive chemotaxin for neurons and may facilitate their contact with their target tissues.34 Furthermore, NGF production is an important physiologic regulator of the neural system, whereby several tissues influence the density and distribution of their sympathetic and sensory innervation.35 In our analysis, NGF was highly expressed in pancreatic cancer cells and its TrkA receptor was mainly present in the perineurium, both of which indicate that NGF may also have the above functions and may be one of the elements essential for nerve growth and contact of nerves to cancer cells, leading to perineural invasion mediated by the TrkA receptor present in the perineurium. NGF and TrkA were also coexpressed in intrapancreatic ganglia, which suggests that NGF might stimulate nerve growth by autocrine and paracrine mechanisms. Obviously, in enlarged pancreatic nerves in pancreatic cancer specimens, additional growth-promoting factors, such as transforming growth factor–alpha, are present8,14-16 that can enhance the growth of pancreatic cancer cells. This may also contribute to the striking affinity between nerve and pancreatic cancer cells and may promote the infiltration and spread of cancer cells along nerves.

Abdominal and back pain are a common symptom of pancreatic cancer. Between 30% and 60% of patients with early, relatively limited disease and 80% of those with advanced disease suffer from pain.36,37 It is thought that abdominal pain in pancreatic cancer patients results from neighboring organ invasion by cancer cells.38 However, there is not a one-to-one correlation of pain with the morphologic presence of cancer cell infiltration into pancreatic nerves, and 24% of patients in whom nerve infiltration by cancer cells was observed did not have pain.2 Therefore, we hypothesize that factors which mediate cancer cell invasion into pancreatic nerves might also be involved in pain generation and that the NGF/TrkA system is a strong candidate in this regard. Previous studies analyzing the influence of NGF on pain generation in nonmalignant disorders have shown that NGF induces thermal hypalgesia39-41 and that it is a regulator of substance P and CGRP expression in sensory neurons, and both substance P and calcitonin gene-related peptide are critical neurotransmitters in the development of pain.42-45 Furthermore, animal models using gene-targeted mutant mice have demonstrated an important role for NGF and its specific receptor, TrkA, in the maturation and function of nociceptive sensory neurons.46,47 Moreover, in four unrelated patients with anhydrosis and insensitivity to pain, mutations in the TrkA gene resulting in the absence of functional TrkA underscore the importance of the NGF/TrkA system in pain generation.48 Inasmuch as NGF plays a significant role in the development of nociceptors and in determining the threshold of their activation,42 experimental and clinical findings strongly indicate that the NGF/TrkA system may function as a mediator of some persistent pain states.

In our analysis, NGF and TrkA mRNA expression levels in pancreatic cancers were significantly higher in patients with pain than in patients without pain. Moreover, there was a significant difference in the intensity of pain depending on whether there was enhanced NGF or TrkA expression. On the basis of our data showing concomitant overexpression of NGF and TrkA in pancreatic cancer and their close relationship to pain generation, we postulate that upregulation of these factors is an important component in pain generation in pancreatic cancer patients, although additional factors are involved in the complex process of pain.

NGF and TrkA can be used in pancreatic cancer as indirect parameters of perineural invasion. As reported before, patients whose tumors exhibit perineural invasion have a poorer prognosis.5,6 Inasmuch as we found no relationship between NGF/TrkA and tumor stage and grading, the addition of both parameters to the tumor stage can identify a subgroup of patients who will have a poorer prognosis compared with other patients in the same tumor stage. This subgroup might benefit from adjuvant therapy and more aggressive surgery, including removal of retroperitoneal nerve tissue along the aorta and the vena cava at tumor resection. Furthermore, the fact that pain was associated with NGF and TrkA levels suggests that medical targeting of NGF/TrkA might decrease pain in pancreatic cancer patients.

In summary, the NGF/TrkA system might be critical for nerve infiltration by pancreatic cancer cells and might also influence pain generation in pancreatic cancer patients.


   ACKNOWLEDGMENTS
 
Supported by the Swiss National Foundation (32-049494.96).


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
1. Friess H, Uhl W, Beger HG, et al: Surgical treatment of pancreatic cancer. Dig Surg11:378-386, 1994

2. Drapiewski JR: Carcinoma of the pancreas: A study of neoplastic invasion of nerves and its possible clinical significance. Am J Clin Pathol14:549-556, 1944

3. Nagakawa G, Mori K, Nakano T, et al: Perineural invasion of carcinoma of the pancreas and biliary tract. Br J Surg80:610-612, 1983

4. Nagakawa T, Kazahara M, Ohta T, et al: Patterns of neural and plexus invasion of human pancreatic cancer and experimental cancer. Int J Pancreatol10:113-119, 1991[Medline]

5. Miyazaki I: Perineural invasion and surgical treatment of the pancreas head cancer. Nippon Geka Gakkai Zasshi98:646-648, 1997[Medline]

6. Takahashi T, Ishikura H, Motohara T, et al: Perineural invasion by ductal adenocarcinoma of the pancreas. J Surg Oncol65:164-170, 1997[Medline]

7. Takahashi T, Ishikura H, Katoh H, et al: Intra-pancreatic, extra-tumoral perineural invasion (nex): An indicator for the presence of retroperitoneal neural plexus invasion by pancreas carcinoma. Acta Pathol Jpn42:99-103, 1992[Medline]

8. Pour PM, Egami H, Takiyama Y: Patterns of growth and metastasis of induced pancreatic cancer in relation to the prognosis and its clinical implications. Gastroenterology100:529-536, 1991[Medline]

9. Kayahara M: Clinicopathological and experimental studies on the spread of pancreatic cancer neural invasion in the extrapancreatic plexus. Gastroenterol Surg21:1363-1372, 1988

10. Gudjonsson B: Cancer of the pancreas: 50 years of surgery. Cancer60:2284-2303, 1987[Medline]

11. Bockman DE, Büchler MW, Beger HG: Morphology of pancreatic cancer: Involvement of nerves, in Beger HG, Büchler MW, Schoenberg M (eds): Cancer of Pancreas. Ulm, Germany, Universitätsverlag Ulm GmbH, 1996, pp 99-108

12. Ernst P: Über das Wachstum und die Verbreitung bösartiger Geschwülste insbesondere des Krebes in den Lymphbahnen der Nerven. Beitr Path Anat7:29-32, 1905

13. Larson DL, Rodin AE, Roberts DK, et al: Perineural lymphatics: Myth or fact. Am J Surg106:488-492, 1966

14. Bockman DE, Büchler MW Beger HG: Interaction of pancreatic ductal carcinoma with nerves leads to nerve damage. Gastroenterology107:219-230, 1994[Medline]

15. Kenmotsu M: Relationship between perineural invasion and expression of neuroendocrine markers and neural cell adhesion molecule in human pancreatic carcinoma. Jpn J Gastroenterol Surg23:2580-2585, 1990

16. Seki H, Tanaka J-I, Sato Y, et al: Neural cell adhesion molecule (NCAM) and perineural invasion in bile duct cancer. J Surg Oncol53:78-83, 1993[Medline]

17. Levi-Montalcini R: The nerve growth factor 35 years later. Science237:1154-1162, 1987[Free Full Text]

18. Oelmann E, Sreter L, Schuller I, et al: Nerve growth factor stimulates clonal growth of human lung cancer cell lines and a human glioblastoma cell line expressing high-affinity nerve growth factor binding sites involving tyrosine kinase signaling. Cancer Res55:2212-2219, 1995[Abstract/Free Full Text]

19. Geldof AA, De Kleijn MA, Rao BR, et al: Nerve growth factor stimulates in vitro invasive capacity of DU 145 human prostatic cancer cells. J Cancer Res Clin Oncol123:107-112, 1997[Medline]

20. Glinsky GV, Ivanova AB, Vinnitsky VB, et al: Nerve growth factor may be the metastasis growth factor. Ann NY Acad Sci496:656-659, 1987[Medline]

21. DeSchryver-Kecskemeti K, Balogh K, Neet KE: Nerve growth factor and the concept of neural-epithelial interactions. Arch Pathol Lab Med111:833-835, 1987[Medline]

22. Klein R, Jing SQ, Nanduri V, et al: The Trk proto-oncogene encodes a receptor for nerve growth factor. Cell65:189-197, 1991[Medline]

23. Kaplan DR, Martin-Zanca D, Parada LF: Tyrosine phosphorylation and tyrosine kinase activity of the Trk proto-oncogene product induced by NGF. Nature350:158-160, 1991[Medline]

24. Sobin LH, Wittekind C: Pancreas. Sobin LH, Wittekind C (eds): TNM classification of malignant tumors (ed 5). New York, NY, Springer-Verlag, 1997, pp 87-90

25. Tanaka A, Watanabe T, Okuno K, et al: Perineural invasion as a predictor of recurrence of gastric cancer. Cancer73:550-555, 1994[Medline]

26. Kashiwagi M, Friess H, Uhl W, et al: Phospholipase A2 isoforms are altered in chronic pancreatitis. Ann Surg227:220-228, 1998[Medline]

27. Friess H, Yamanaka Y, Büchler MWet al: Enhanced expression of transforming growth factor beta isoforms in pancreatic cancer correlates with decreased survival. Gastroenterology105:1846-1856, 1993[Medline]

28. Guo XZ, Friess H, Graber HU, et al: KAI1 expression is up-regulated in early pancreatic cancer and decreased in the presence of metastasis. Cancer Res56:4876-4880, 1996[Abstract/Free Full Text]

29. Yamanaka Y, Friess H, Büchler MWet al: Synthesis and expression of transforming growth factor beta 1, beta 2 and beta 3 in endocrine and exocrine pancreas. Diabetes42:746-756, 1993[Abstract]

30. Liu HM, Yang HB, Chen RM, et al: Expression of basic fibroblastic growth factor, nerve growth factor, platelet-derived growth factor and transforming growth factor-beta in human brain abscess. Acta Neuropathol88:143-150, 1994[Medline]

31. Tanaka T, Hiyama E, Sugimoto T, et al: TrkA gene expression in neuroblastoma: The clinical significance of an immunohistochemical study. Cancer76:1086-1095, 1995[Medline]

32. Lu Z Friess H, Graber HU, et al: The presence of two signaling TGF-beta receptors in human pancreatic cancer correlates with advanced tumor stage. Dig Dis Sci42:2054-2063, 1997[Medline]

33. Yamamoto M, Iseki S: Nerve growth factor receptor (NGFR)-like immunoreactivity in the perineural cells of the adult rat peripheral cells: Normal appearance and response to nerve section. Kaibogaku Zasshi67:642-649, 1992[Medline]

34. Gundersen RW, Barrett JN: Characterization of the turning response of dorsal-root neurites toward nerve growth factor. J Cell Biol87:546-554, 1980[Abstract/Free Full Text]

35. Paul AB, Grant ES, Habib FK: The expression and localization of beta-nerve growth factor (NGF) in benign and malignant human prostate tissues: Relationship to neuroendocrine differentiation. Br J Cancer74:1990-1996, 1996[Medline]

36. Greenwald HP, Bonica JJ, Bergner M: The prevalence of pain in four cancers. Cancer60:2563-2569, 1987[Medline]

37. Krech RL, Walsh D: Symptoms of pancreatic cancer. J Pain Symptom Manage6:360-367, 1991[Medline]

38. Caraceni A, Portenoy RK: Pain management in patients with pancreatic carcinoma. Cancer78:639-653, 1996 (suppl) [Medline]

39. McMahon SB: NGF as a mediator of inflammation pain. Philos Trans R Soc Lond B Biol Sci351:431-440, 1996[Medline]

40. Lewin GR, Rueff A, Mendell LM: Peripheral and central mechanisms of NGF-induced hyperalgesia. Eur J Neurosci6:1903-1912, 1994[Medline]

41. Woolf CJ: Phenotypic modification of primary sensory neurons: The role of nerve growth factor in the production of persistent pain. Philos Trans R Soc Lond B Biol Sci351:441-448, 1996[Medline]

42. Malcangio M, Garrett NE, Cruwys S, et al: Nerve growth factor and neurotrophin-3-induced changes in nociceptive threshold and the release of substance P from the rat isolated spinal cord. J Neurosci17:8459-8467, 1997[Abstract/Free Full Text]

43. Otten U, Goedert M, Mayer N, et al: Requirement of nerve growth factor for development of substance P-containing sensory neurons. Nature287:158-159, 1980[Medline]

44. Joseph A, Markenson MD Mechanisms of chronic pain. Am J Med 101:6s-18s, 1996

45. Lindsay RM, Harmar AJ: Nerve growth factor regulates expression of neuropeptide genes in adult sensory neurons. Nature337:362-364, 1989[Medline]

46. Crowley C, Spencer SD, Nishimura MCet al: Mice lacking nerve growth factor display perinatal loss of sensory and sympathetic neurons yet develop basal forebrain cholinergic neurons. Cell76:1001-1002, 1994[Medline]

47. Smeyne RJ, Schnapp A, Long LK, et al: Severe sensory and sympathetic neuropathies in mice carrying a disrupted Trk/NGF receptor gene. Nature368:246-249, 1994[Medline]

48. Indo Y, Tsuruta M, Hayashida Y, et al: Mutations in the TrkA/NGF receptor gene in patients with congenital insensitivity to pain with anhidrosis. Nat Genet13:485-488, 1996[Medline]

Submitted October 30, 1998; accepted March 29, 1999.




This article has been cited by other articles:


Home page
 Cancer Res.Home page
B. J. Swanson, K. M. McDermott, P. K. Singh, J. P. Eggers, P. R. Crocker, and M. A. Hollingsworth
MUC1 Is a Counter-Receptor for Myelin-Associated Glycoprotein (Siglec-4a) and Their Interaction Contributes to Adhesion in Pancreatic Cancer Perineural Invasion
Cancer Res., November 1, 2007; 67(21): 10222 - 10229.
[Abstract] [Full Text] [PDF]

Home page
 J. Pharmacol. Exp. Ther.Home page
K. D. Wild, D. Bian, D. Zhu, J. Davis, A. W. Bannon, T. J. Zhang, and J.-C. Louis
Antibodies to Nerve Growth Factor Reverse Established Tactile Allodynia in Rodent Models of Neuropathic Pain without Tolerance
J. Pharmacol. Exp. Ther., July 1, 2007; 322(1): 282 - 287.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
N. Koide, T. Yamada, R. Shibata, T. Mori, M. Fukuma, K. Yamazaki, K. Aiura, M. Shimazu, S. Hirohashi, Y. Nimura, et al.
Establishment of Perineural Invasion Models and Analysis of Gene Expression Revealed an Invariant Chain (CD74) as a Possible Molecule Involved in Perineural Invasion in Pancreatic Cancer
Clin. Cancer Res., April 15, 2006; 12(8): 2419 - 2426.
[Abstract] [Full Text] [PDF]

Home page
 J. Pharmacol. Exp. Ther.Home page
J. J. Watson, M. S. Fahey, E. van den Worm, F. Engels, F. P. Nijkamp, P. Stroemer, S. McMahon, S. J. Allen, and D. Dawbarn
TrkAd5: A Novel Therapeutic Agent for Treatment of Inflammatory Pain and Asthma
J. Pharmacol. Exp. Ther., March 1, 2006; 316(3): 1122 - 1129.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
G. M. Sclabas, S. Fujioka, C. Schmidt, Z. Li, W. A.I. Frederick, W. Yang, K. Yokoi, D. B. Evans, J. L. Abbruzzese, K. R. Hess, et al.
Overexpression of Tropomysin-Related Kinase B in Metastatic Human Pancreatic Cancer Cells
Clin. Cancer Res., January 15, 2005; 11(2): 440 - 449.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
C. Veit, F. Genze, A. Menke, S. Hoeffert, T. M. Gress, P. Gierschik, and K. Giehl
Activation of Phosphatidylinositol 3-Kinase and Extracellular Signal-Regulated Kinase Is Required for Glial Cell Line-Derived Neurotrophic Factor-Induced Migration and Invasion of Pancreatic Carcinoma Cells
Cancer Res., August 1, 2004; 64(15): 5291 - 5300.
[Abstract] [Full Text] [PDF]

Home page
 Hum ReprodHome page
V. Anaf, I. El Nakadi, Ph. Simon, J. Van de Stadt, I. Fayt, Th. Simonart, and J.-C. Noel
Preferential infiltration of large bowel endometriosis along the nerves of the colon
Hum. Reprod., April 1, 2004; 19(4): 996 - 1002.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
K. Ketterer, S. Rao, H. Friess, J. Weiss, M. W. Buchler, and M. Korc
Reverse Transcription-PCR Analysis of Laser-Captured Cells Points to Potential Paracrine and Autocrine Actions of Neurotrophins in Pancreatic Cancer
Clin. Cancer Res., November 1, 2003; 9(14): 5127 - 5136.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
S. J. Miknyoczki, W. Wan, H. Chang, P. Dobrzanski, B. A. Ruggeri, C. A. Dionne, and K. Buchkovich
The Neurotrophin-Trk Receptor Axes Are Critical for the Growth and Progression of Human Prostatic Carcinoma and Pancreatic Ductal Adenocarcinoma Xenografts in Nude Mice
Clin. Cancer Res., June 1, 2002; 8(6): 1924 - 1931.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
Z.-w. Zhu, H. Friess, L. Wang, T. Bogardus, M. Korc, J. Kleeff, and M. W. Büchler
Nerve Growth Factor Exerts Differential Effects on the Growth of Human Pancreatic Cancer Cells
Clin. Cancer Res., January 1, 2001; 7(1): 105 - 112.
[Abstract] [Full Text]

Home page
 GutHome page
F F di Mola, H Friess, Z W Zhu, A Koliopanos, T Bley, P Di Sebastiano, P Innocenti, A Zimmermann, and M W Buchler
Nerve growth factor and Trk high affinity receptor (TrkA) gene expression in inflammatory bowel disease
Gut, May 1, 2000; 46(5): 670 - 679.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a colleague
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Save to my personal folders
Right arrow Download to citation manager
Right arrowRights & Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhu, Z.
Right arrow Articles by Büchler, M. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhu, Z.
Right arrow Articles by Büchler, M. W.