Skip to main content
NCBI home page
International Journal of Experimental Pathology logoLink to International Journal of Experimental Pathology
. 2010 Aug;91(4):357–367. doi: 10.1111/j.1365-2613.2010.00721.x

Establishment and characterization of a murine xenograft model of appendiceal mucinous adenocarcinoma

Arun A Mavanur *, Vamsi Parimi *, Mark O’Malley *, Marina Nikiforova †, David L Bartlett *, Jon M Davison †
PMCID: PMC2962894  PMID: 20586814

Abstract

We describe the clinical, pathologic and molecular characteristics of a xenograft model of metastatic mucinous appendiceal adenocarcinoma. Tumours from patients with mucinous appendiceal neoplasms were implanted in nude mice and observed for evidence of intraperitoneal tumour growth. Morphologic and immunohistochemical features, temporal growth characteristics relative to controls, and loss of heterozygosity (LOH) at multiple chromosomal alleles were assessed in a successfully engrafted tumour. Two of seventeen implanted tumours successfully engrafted and only one mucinous adenocarcinoma propagated throughout the course of the study. The successful xenograft is morphologically similar to the original tumour, produces abundant extracellular mucin and exhibits non-invasive growth on peritoneal surfaces. The temporal growth characteristics of the xenograft tumour relative to controls reveal that tumour burden can be followed indirectly by measuring the weight or abdominal girth of engrafted animals. The cytokeratin, mucin core protein, CDX2, Ki-67 and p53 expression patterns are identical in the xenograft and resected tumour and are consistent with the expected pattern of protein expression for mucinous adenocarcinoma of the appendix. LOH was found in 1 of 10 informative chromosomal loci (chromosome 10p23) in xenograft tumour cells. Although we were unable to engraft a low-grade appendiceal mucinous neoplasm, the engrafted adenocarcinoma will be useful for future evaluation of novel therapeutic strategies directed at mucinous appendiceal adenocarcinoma and evaluation of strategies for treating widespread, bulky, mucinous peritoneal surface neoplasms. Xenograft tumour enrichment can facilitate molecular studies of appendiceal epithelial neoplasia.

Keywords: adenocarcinoma, appendix, loss of heterozygosity, mouse model, pseudomyxoma peritonei, xenograft

Introduction

Appendiceal mucinous neoplasms are rare intestinal tumours which are known to be the most common cause of ‘pseudomyxoma peritonei’ (PMP), a term which historically has referred to a condition characterized macroscopically by numerous, often bulky, mucinous tumour implants on the visceral and parietal peritoneal surfaces and, in some cases, the production of significant amounts of extracellular mucin (Young et al. 1991; Ronnett et al. 1995a,b, 1997; Szych et al. 1999; Mukherjee et al. 2004). It has a reported incidence of 2 out of every 10,000 laparotomies (Smeenk et al. 2008). The extracellular mucin may accumulate enough to cause mucinous ascites and abdominal distention, resulting in a clinical appearance that has acquired the colloquial description ‘jelly belly’. Ultimately, the widespread tumour implants and mucin result in bowel obstruction and death due to progressive bowel dysfunction and starvation (Sugarbaker 2006).

A pathologic spectrum of mucinous appendiceal neoplasms progresses to widespread, mucinous peritoneal disease. This spectrum includes low-grade appendiceal mucinous neoplasms (LAMN), mucinous adenocarcinoma, and tumours exhibiting intermediate features (Misdraji et al. 2003). LAMN are distinguished from mucinous adenocarcinomas on the basis of several histopathologic differences, including cytologic and architectural atypia, degree of cellularity, presence of parenchymal invasion and presence of nodal or distant hematogenous metastases (Ronnett et al. 1995a,b; Carr & Sobin 1996; Misdraji et al. 2003; Bradley et al. 2006; Pai et al. 2009).

The distinction between LAMN and adenocarcinoma has prognostic significance. When treated with aggressive cytoreductive surgery and intraperitoneal chemotherapy, metastatic LAMN (also referred to as ‘disseminated peritoneal adenomucinosis’ and ‘low-grade mucinous carcinoma peritonei’) have a more favourable prognosis than metastatic mucinous adenocarcinomas (also referred to as ‘peritoneal mucinous carcinomatosis’ and ‘high-grade mucinous carcinoma peritonei’) (Ronnett et al. 1995a,b, 2001; Miner et al. 2005; Stewart et al. 2006; Smeenk et al. 2007; Baratti et al. 2008).

It has also been observed that mucinous peritoneal surface tumours arising from the appendix, as a group, are clinically more indolent than peritoneal carcinomatosis caused by colorectal adenocarcinoma (Sugarbaker & Jablonski 1995). This observation cautions against attempts to categorically apply treatment strategies developed for colorectal adenocarcinoma to appendiceal tumours. Appendiceal tumours are likely to require tailored approaches to treatment in order to maximize therapeutic efficacy.

An increasing amount of scholarship devoted to the topic of PMP indicates that worldwide clinical experience with this disease is growing (Smeenk et al. 2008). However, the relative rarity of this condition precludes large, randomized clinical trials – a circumstance which demands the use of other experimental strategies. Appropriate animal models could play a pivotal role in this context. Tumour xenograft models have played an important part in the evaluation of novel therapeutic strategies for human tumours as a complement to transgenic models. They are also an essential experimental system for the study of tumours which are not currently amenable to transgenic techniques.

A suitable xenograft model for evaluating novel therapeutic approaches to treating widespread mucinous neoplasms would accurately recapitulate essential pathologic features of the disease, exhibit stability over generations, and exhibit quantifiable clinical signs of tumour growth in all engrafted animals to allow indirect measurement of tumour burden.

Xenograft technology has also facilitated the study of tumour biology and genetics. Growth as a xenograft has the potential to substantially enrich tumour cells relative to contaminating normal stromal cells which tend to mask detection of genetic abnormalities in fresh tumour samples. So called ‘xenograft enrichment’ facilitated the study of pancreatic adenocarcinoma, a human tumour characterized by an abundance of (desmoplastic) stromal elements and proportionally little tumour epithelium (Hahn et al. 1995). In fact, the majority of any mucinous appendiceal tumour is comprised of mucin, normal fibroblasts, inflammatory cells and other stromal elements and only rare fragments of neoplastic epithelium. Xenograft enrichment offers promise for the study of appendiceal neoplasia as well. Novel strategies for treatment may follow from an improved understanding of the molecular pathogenesis of appendiceal neoplasia.

To address the urgent need for appropriate experimental models of PMP and to develop methods for the engraftment of human appendiceal mucinous tumours in mice for molecular studies, we report the successful engraftment of an appendiceal mucinous adenocarcinoma. We show that the xenograft is morphologically and immunohistochemically identical to the original human tumour and describe the distinctive pathologic features which recommend it as a potentially useful model of abundantly mucinous, bulky, superficial, intra-abdominal epithelial neoplasms. Moreover, we provide the first experimental description of the time course of tumour growth in inoculated animals as compared to controls. Our description of these clinical characteristics will provide much needed information for planning therapeutic intervention trials with this model and serves as a standard for comparison with the clinical characteristics of other models. The tumour tissue obtained from this model is enriched in human tumour DNA which was used for evaluation of allelic imbalances at multiple chromosomal loci. This xenograft line will serve as a model system to explore therapeutic options for the almost invariably fatal mucinous appendiceal adenocarcinoma. The experience we share will hopefully lead to improved techniques for xenografting appendiceal mucinous tumours for future experimental therapeutic and molecular studies.

Methods

Animals

All studies were performed with homozygous nude mice (CrTac:NCr-Foxn1nu) obtained from Taconic (Tarrytown, NY, USA) using procedures that were approved by the University of Pittsburgh Animal Care and Use Committee.

Patient samples

Surgically resected tumour samples were obtained from preoperatively consented patients diagnosed with metastatic appendiceal mucinous neoplasms. Experimental procedures were approved by the University of Pittsburgh Institutional Review Board (Protocol 02-077).

Tumour grafting and passaging

Immediately following surgical resection, tumour samples were transported to the laboratory for implantation into recipient mice. Individual 0.5 cm3 tumour blocks from patient tumour samples were immersed in phosphate-buffered saline (PBS) containing 100 U/ml penicillin and 100 μg/ml streptomycin for 3 h. These samples were implanted in the peritoneal cavity of nude mice by means of a 1 cm transverse abdominal incision performed under aseptic conditions. Mice were maintained on 2% isoflurane anaesthesia via nose cone until completion of the procedure. Following implantation, the incision was closed with 4–0 silk sutures. Three to five mice were implanted with tumour samples from each patient.

After the surgical implantation of the human tumour samples, the mice were raised in a pathogen-free environment and allowed food and water ad libitum. After tumour implantation, mice were observed weekly for evidence of intra-abdominal tumour growth (abdominal distention and ascites). Mice were sacrificed after approximately 8 weeks for necropsy in order to evaluate intraperitoneal tumour growth and to harvest viable, engrafted tumours for subsequent passage.

Engrafted tumour was passaged to a subsequent generation of nude mice by preparing a disaggregated tumour suspension from approximately 1 ml of gelatinous xenograft tumour in 10 ml of PBS by gently drawing suspended tumour tissue into a 5 ml syringe through an 18-gauge needle under sterile conditions until tissue aggregates were dissociated. An aliquot of the suspension (1.5–2 ml) was then injected into the peritoneal cavity of multiple recipient mice (n = 5). Care was taken prior to each passage to aggregate tumour samples from multiple mice (n = 5) before transferring tumour to the subsequent generation of recipient mice in order to prevent clonal divergence in subsequent generations.

In order to evaluate the rate of growth of the xenograft, tumour from sixth generation mice was suspended in PBS as described above and 1.5 ml was injected intraperitoneally into ten nude mice 4–6 weeks in age. Media alone was introduced into the peritoneal cavity of 10 age-matched controls. Weights and abdominal girth were measured weekly. After 60 days, all animals were sacrificed and the abdominal cavity was examined for the presence of mucinous ascites.

Fresh tumour tissue from generation four was frozen to determine if it could be stored for future passage. Approximately 3 ml of mucinous xenograft tumour tissue was suspended in an equal volume of Dulbecco’s Modified Eagle’s Medium (DMEM) with 20% fetal bovine serum and 10% dimethyl sulphoxide. The suspended tissue was placed into 1 ml aliquots and slow frozen at the rate of 0.5 °C/min to −20 °C and then at 1 °C/min to −80 °C. Frozen tumour was thawed after 6 months and resuspended in an equal volume of PBS. One millilitre of the resuspended solution was injected intraperitoneally into nude mice. These mice were observed for 60 days, sacrificed and examined for evidence of mucinous ascites.

Gross pathologic and histologic evaluation

After the first, second and sixth tumour passage, complete necropsy was performed on five mice showing evidence of abdominal distension. The abdominal and thoracic organ blocks were examined grossly and representative tissues were fixed in formalin and embedded in paraffin.

The following gross features were evaluated at necropsy: extent of gross spread (limited to site of injection, sites of peritoneal involvement, spread above diaphragm); formation of solid masses or gelatinous ascites; gross invasion into organs or surface involvement only, evidence of hematogenous metastases.

Five micron thick sections were cut for histologic evaluation of tumour implants from paraffin embedded tissues and stained with haematoxylin and eosin using standard protocols. Histologic characteristics which are known to correlate with the clinical behaviour of appendiceal mucinous neoplasms include: abundance of epithelium (relative to extracellular mucin), architectural features of the tumour epithelium (simple strips of epithelium, villiform projections, cribriform structures, solid sheets, single cells, desmoplastic invasion, angiolymphatic invasion, lymph node involvement, parenchymal invasion) and cytologic features (presence of nuclear size variation, nuclear enlargement/elongation/hyperchromasia, large prominent nucleoli, mitotic activity, necrosis). These histologic characteristics were evaluated in the resected tumour as well as the xenografted tumour after the first, second and sixth passages and the tumour was classified according to previously established criteria (Ronnett et al. 1995a,b; Misdraji et al. 2003).

Immunohistochemistry

The original tumour and selected samples from passage two xenografts were evaluated for the expression of the following antigens by immunohistochemistry: cytokeratin 7 (Dako, Carpinteria, CA, USA; clone OV-TL 12/30, 1:100), cytokeratin 20 (Dako, clone Ks20.8, 1:20), CDX2 (Biogenex, San Ramon, CA, USA; clone CDX2-88, 1:100), MUC-2 (Neomarkers, clone M53, 1:100), MUC-5AC (Neomarkers, Fremont, CA, USA; clone 45M1, 1:25), MUC-6 (Vector Laboratories, Burlingame, CA, USA; clone CLH5, 1:25), Ki-67 (Dako, clone MIB-1, 1:50) and p53 (Dako, clone DO-7, 1:100). Staining was performed on 4-μm paraffin sections with microwave pretreatment in citrate buffer pH 6.0 for 10 min. The Universal Impress kit (Vector Laboratories) was used for antibody detection with DAB chromogen (Vector Laboratories) and haematoxylin counterstain. All staining was performed by hand with the exception of staining for MUC6 which was performed on the Ventana Benchmark XT automated system. Additional details of staining protocols are available upon request.

Loss of heterozygosity and K-ras mutational analysis

To compare patterns of chromosomal allelic imbalance (loss of heterozygosity; LOH) between the xenograft and primary tumour, tissue from the initial xenograft and passage 2 were manually dissected from 4-μm thick unstained paraffin sections using a high-resolution stereomicroscope, utilizing an adjacent H&E stained section as a guide. Normal splenic tissue from the patient as well as tissue from the resected tumour was obtained in a similar fashion. DNA was isolated from each tissue sample (xenograft tumour, original tumour, spleen) using the DNeasy tissue kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. A panel of 14 polymorphic microsatellite markers was used to evaluate LOH at different chromosomal loci. These included chromosomes 1p36 (D1S171, D1S407), 3p25 (D3S1516), 5q23 (D5S615, D5S1384), 7q22 (D7S1530), 7q31 (D7S486), 9p21 (D9S251), 9q22 (D9S252), 9p22 (D9S1748), 10q23 (D10S520, D10S1173), 17p13 (D17S1289) and 18q21 (D18S487). PCR amplification was performed using fluorescently labelled primers and the products of amplification were detected by capillary gel electrophoresis on ABI3730 (Applied Biosystems, Foster City, CA, USA). The relative fluorescence values (peak heights) were obtained for individual alleles and the ratio of peaks was calculated using genemapper software v3.2 (Applied Biosystems). Only alleles which show two distinct PCR products of different molecular weight (i.e. heterozygous) in the normal tissue obtained from the patient were regarded as informative. At informative loci, LOH was determined by a comparison of allele peak height ratios between tumour samples and normal tissue using the following formula (NA/NB)/(TA/TB); NA = peak height of allele A (larger molecular weight) in normal tissue; NB = peak height of allele B in normal tissue; TA = peak height of allele A (larger molecular weight) in tumour sample; TB = peak height of allele B in tumour sample. When the ratio was <0.5 or >2.0, it was considered evidence of LOH for a given locus as described previously (Marsh et al. 2003). In practical terms, this is usually reflected in a marked decrease in the peak height of one PCR product relative to the other in the tumour sample. To be regarded as LOH, we required that two independent tumour samples had to meet peak height ratio criteria in order to avoid the possibility of false positive results.

Detection of codon 12 or 13 mutation in exon 1 of the K-ras gene was performed by direct nucleotide sequencing using the BigDye Terminator Kit on ABI3130 (Applied Biosystems) as described previously and according to manufacturers specifications (Yousem et al. 2008). The sequencing results were analysed for the presence of mutations with Mutation Surveyor v.3.01 (softgenetics). Both sense and antisense sequencing was performed to assure the detection of heterozygous mutations in the K-ras gene.

Statistical analysis

Mean values of weight and girth were compared by Student’s t-test using stata, version 10.1 software (StataCorp, College Station, TX, USA). Values of P < 0.05 were considered significant.

Results

Initial engraftment of resected human tumours

Successful engraftment of initially implanted tumours occurred within 8 weeks in 2 of 17 (12%) of resected mucinous appendiceal tumours after implantation in the peritoneal cavity of nude mice. Both of the tumours which initially engrafted were appendiceal adenocarcinomas which had spread to the peritoneal cavity of the patients from whom they were derived. One implanted tumour (PMP-754) initially resulted in widespread intraperitoneal tumour formation and mucinous ascites in recipient mice, resembling the macroscopic findings in human patients (Figure 1). The other (PMP-731) initially formed a localized mass near the site of intraperitoneal implantation, but did not passage and was not further studied (data not shown).

Figure 1.

Figure 1

Open in a new tab

Pathology and immunohistochemistry of PMP-754 with comparison to original human tumour. (a) Mouse with intraperitoneal xenograft PMP-754 (day 58). There is abundant mucinous ascites and gelatinous masses filling the entire peritoneal cavity in this mouse with a distended abdomen. (b) Organ block containing liver, gallbladder, pancreas, stomach, duodenum and spleen from day 58 mouse with intraperitoneal xenograft. Gelatinous tumour nodules are present on the serosal surface of all organs. (c) Segment of bowel from same mouse as in (b) demonstrating mucinous nodules on the small bowel mesentery and adherent to the small bowel serosa. (d) Histologic section of a superficial peritoneal deposit representing the original human tumour. This section shows cribriform glands with severe cytologic dysplasia and stromal reaction with minimal fibrosis. (e, f) Histologic sections of the xenograft tumour demonstrating cribriform glandular architecture with severe cytologic dysplasia and mitotic activity similar to the original human tumour. This histologic appearance was consistent through six passages. (g) Histologic section demonstrating tumour involvement of the serosal surface of the liver. No evidence of macroscopic or microscopic parenchymal invasion is seen. (h, i) Immunohistochemical staining for Ki-67. The original tumour (h) and the xenograft (i) show an elevated Ki-67 proliferative index. (j, k) Immunohistochemical staining for p53. Heterogeneous nuclear staining is seen in the majority of tumour cell nuclei of the original tumour (j) and the xenograft (k).

Clinical and pathologic characteristics of the successfully engrafted mucinous adenocarcinoma

The patient was a 58-year-old female who underwent an appendectomy in 2007 at an outside institution where she was found to have extensive mucinous carcinomatosis. The appendix was involved by a ruptured adenocarcinoma in situ lacking definitive invasion. Four months later she underwent cytoreductive surgery at our institution. Pathologic evaluation of the resected specimens identified tumour deposits in the omentum, on the surface of the ovaries and fallopian tubes bilaterally, the serosal surface of the spleen, diaphragm, gallbladder and parietal peritoneal surfaces. There was no evidence of parenchymal organ invasion. Histologically, the peritoneal tumour deposits were characterized by moderately abundant glandular epithelium forming strips and cribriform structures and abundant extracellular mucin production (Figure 1). The glands were comprised of cells with moderate to severe cytologic atypia (marked nuclear pleomorphism, enlarged nucleoli, loss of nuclear polarity and elevated mitotic activity, Figure 1). These histologic features are diagnostic of mucinous appendiceal adenocarcinoma (or peritoneal mucinous carcinomatosis, PMCA) according to previously published criteria (Ronnett et al. 1995a,b; Misdraji et al. 2003).

Gross morphology and histologic characteristics of PMP-754

We performed a detailed necropsy on a total of 10 mice after passage 2 and passage 6 of PMP-754 in order to evaluate pathologic features of tumour spread in mice. Findings were uniform throughout the group when evaluated at 60 days after intraperitoneal injection of tumour. At this time point, the mice have a noticeably distended abdomen. When the abdominal cavity is opened, abundant mucinous tumour is encountered, encasing all abdominal organs (Figure 1). We observed gelatinous tumours loosely adherent to serosal surfaces of all intraabdominal organs as well as gelatinous ascites. Localized masses were occasionally observed. We did not observe tumour spread above the diaphragm in intraperitoneally injected mice. Neither direct parenchymal invasion nor evidence of hematogenous metastases was seen grossly (Figure 1). When examined histologically, PMP-754 is an epithelial tumour which produces abundant extracellular mucin (Figure 1). There is no evidence of microscopic invasion of parenchymal organs or hematogenous metastases in sections taken from the grossly normal liver, lungs and spleen. The tumour grew freely in the abdominal cavity and on the serosal surfaces of intra-abdominal organs. In contrast to the human tumour, the mouse tumour at 60 days did adhere to serosal surfaces of the intestine – a consequence of the massive amount of tumour which is present at this time point. The gelatinous tumour mass was infiltrated by stromal cells and blood vessels, presumably from the host. In contrast to the human tumour which exhibited delicate bands of fibrosis within tumour deposits (a typical feature of low grade tumours as well as mucinous adenocarcinoma in human tumours), fibrosis was not significant in the xenograft. This tumour shows severe cytologic and architectural atypia and elevated mitotic rate nearly identical to the resected patient tumour and was classified as mucinous adenocarcinoma (or PMCA, Figure 1) according to previously published criteria (Ronnett et al. 1995a,b; Misdraji et al. 2003).

Engraftment frequency during passage and growth characteristics of PMP-754

In order to evaluate the engraftment frequency and time course of tumour formation, a group of mice were injected intraperitoneally with PMP-754 tumour cells suspended in PBS. Controls were injected with PBS alone. By 38 days postinjection, the mean weight of tumour bearing mice (25.9 g, range 23.1–29.4) was significantly greater than that of controls (22.2 g, range 21.4–23.5) (P < 0.001, Figure 2). At later time points when larger differences in mean weight are measured, we saw no overlap in the range of weights between tumour bearing mice and controls. By 48 days postinjection, the mean abdominal girth of tumour engrafted mice (88.9 mm, range 75–97 mm) was significantly different from that of controls (69.0 mm, range 69–71 mm) (P< 0.0001, Figure 2).

Figure 2.

Figure 2

Open in a new tab

Change in weight and abdominal girth in PMP-754 engrafted animals vs. controls. Figures demonstrate change in weight (a) and girth (b) over time in PMP-754 bearing mice (▪) and controls (▴). No significant differences in weight (a) are present until 38 days after intra-peritoneal tumour injection (P < 0.001 for this time point). Likewise, no significant difference in abdominal girth (b) is measured until day 48 (P < 0.0001 for this time point). Error bars represent standard deviation.

All 10 mice that received peritoneal injections of PMP-754 had mucinous ascites on examination of the peritoneal cavity at 58 days, indicating a 100% frequency of engraftment into nude mice when fresh tumour is transferred. However, only 40% (two of five) of mice developed mucinous ascites by 60 days when stored, frozen tumour was resuspended and introduced into the peritoneal cavity. PMP-754 has been passaged through seven generations to date.

Expression of p53, Ki-67 and markers of intestinal differentiation

In order to confirm the appendiceal origin of the xenograft and compare the immunophenotype of xenograft PMP-754 with the original tumour, we evaluated expression of p53, Ki-67, cytokeratin 7, cytokeratin 20, CDX2, MUC-2, MUC-5AC and MUC-6 by immunohistochemistry. PMP-754 and the original human tumour displayed an identical pattern of cytokeratin, mucin core protein and CDX2 expression which is typical of appendiceal mucinous tumours (Figure 1, Table 1) (Yajima et al. 2005; Nonaka et al. 2006; Yoon et al. 2009). Both the xenograft and the resected tumour exhibited a high Ki-67 proliferative index and displayed nuclear p53 immunoreactivity with heterogeneous nuclear intensity (ranging from weak to intense, Figure 1, Table 1).

Table 1.

Results of immunohistochemistry on the original tumour and PMP-754

Original tumour PMP-754
Cytokeratin 20 + +
Cytokeratin 7 − −
CDX2 + +
MUC-2 + +
MUC-5AC Focal + Focal +
MUC-6 − −
P53 + +
Ki-67 +, 30% +, 50%
Open in a new tab

Loss of heterozygosity and K-ras mutation analysis for PMP-754

Ten of 14 chromosomal loci were informative for evaluation of LOH. By the strict criteria we employed, LOH was not detected at any locus in the original human tumour specimen. However, LOH was detected at one locus (D10S1173, chromosome 10p23) in the xenograft tumour samples (Figure 3). It is noteworthy that the original tumour shows a reduced peak height for the same size PCR product which is almost entirely absent in the xenograft tumour (Figure 3). However, the magnitude of peak height reduction in the resected tumour specimen is insufficient to meet our criteria for LOH. Although abnormal nuclear accumulation of p53 protein was detected by immunohistochemistry, LOH at chromosome 17p13 was not detected (Figure 3). The genomic sequence of the K-ras gene (KRAS2), spanning the frequently mutated codons 12 and 13 was normal in the original human tumour and in the xenograft (data not shown).

Figure 3.

Figure 3

Open in a new tab

LOH analysis at D10S1173 and D17S1289. In these partial electropherograms, the horizontal axis represents molecular weight of the PCR product and the vertical axis represents fluorescence intensity. Note that the scales are different for each electropherogram. Two alleles of distinct molecular weight are present in the normal tissue DNA from the patient at both loci. For D10S1173, the xenograft tumour shows a marked reduction of fluorescence intensity for the lower molecular weight PCR product (asterisk) indicating allelic imbalance at this chromosomal locus. A slight decrement in fluorescence intensity is noted in the original human tumour sample, though of insufficient magnitude to confidently regard it as LOH. This difference may be due to contaminating normal cells which tends to mask the presence of allelic imbalance. Allelic imbalance is neither detected in the original tumour nor the xenograft at D17S1289.

Discussion

The histopathologic features of this xenograft are nearly identical to those of the resected mucinous adenocarcinoma from which it was derived. Similar to patients with metastatic appendiceal mucinous adenocarcinoma, engrafted mice develop progressive intra-abdominal disease and eventually require euthanasia.

Although this adenocarcinoma has cytologic and architectural atypia exceeding that which is seen in LAMN, it has some features in common with LAMN: it produces abundant extra-cellular mucin leading to the uniform development of mucinous ascites; it exclusively involves serosal surfaces and does not exhibit parenchymal invasion; lastly, there is no evidence of nodal or hematogenous metastases. Given these similarities, it will serve as a model of a highly mucinous peritoneal surface malignancy. These tumours present unique clinical challenges due to the bulky, non-localized nature of disease and the presence of an acellular mucinous matrix which surrounds most viable tumour cells. This model will provide an experimental system in which to evaluate strategies to overcome these obstacles to effective therapy.

Nevertheless, in spite of having some similarities to LAMN, we recognize that this tumour model represents the more clinically aggressive appendiceal adenocarcinoma and is likely to be biologically different from LAMN due to the greater degree of morphologic atypia seen in this mucinous tumour and its high proliferative index. Appendiceal adenocarcinoma responds poorly to current multimodality therapy and will require novel chemotherapeutic strategies which can be explored utilizing this animal model.

The question naturally arises, ‘Is this a model of PMP?’ Unfortunately, this question is difficult to answer satisfactorily given that the term PMP is not well defined. Recently, because of observed differences in clinical behaviour of LAMN and mucinous adenocarcinoma of the appendix, Ronnett et al. recommended that PMP be defined as the clinical manifestation of metastatic low-grade appendiceal mucinous neoplasms (or disseminated peritoneal adenomucinosis, DPAM) (Ronnett et al. 1995a,b;). The merit of this restricted definition of PMP is that it encompasses metastatic tumours with relatively indolent rate of progression which can be treated successfully with aggressive cytoreductive surgery with or without intraperitoneal chemotherapy and excludes cases for which this aggressive therapeutic approach offers less benefit (i.e. frank adenocarcinoma). In this sense, our engrafted adenocarcinoma is not a model of PMP because it is derived from a metastatic appendiceal adenocarcinoma.

However, this xenografted tumour does model ‘PMP’, if one takes ‘PMP’ in a broader context. PMP is a diagnosis which has historically referred to mucinous neoplasms representing a pathologic spectrum from low-grade mucinous neoplasms to adenocarcinoma all sharing in common the fact that they are widely metastatic, macroscopically predominantly confined to peritoneal surfaces, produce abundant extracellular mucinous matrix and, in many cases, arise from primary appendiceal tumours. Ultimately, this question is of little clinical relevance, in our opinion, because it is the pathologic diagnosis explicitly identifying the primary tumour, its grade, stage, pattern of organ involvement determining its potential for complete resection, molecular and biologic properties, etc. which determines the therapeutic approach and prognosis regardless of the macroscopic appearance of the tumour and whether it fits within a narrow or broad conception of the term ‘PMP’.

One characteristic pathologic property which was not seen in histologic sections taken of the xenograft tumours was the prominent fibrosis which is usually elicited by dissecting and free mucin within the peritoneal cavity in human tumours. This pathologic feature was also absent in xenografts of appendiceal adenocarcinoma described in the only other report of successful engraftment of appendiceal adenocarcinoma (Flatmark et al. 2007). Fibrosis is one potential cause of bowel dysfunction in patients with advanced peritoneal disease. Both PMP-754 and the previously reported xenograft of mucinous appendiceal adenocarcinoma were established in athymic (nude) mice. In spite of extensive tumour deposition, there is an absence of fibrosis which raises intriguing questions about the role of T cells in regulating tumour- or mucin-induced peritoneal fibrosis.

We have demonstrated that PMP-754 transfers with 100% success when mice are inoculated with fresh tumour suspension. Weight is significantly different in tumour inoculated mice vs. age-matched controls by 38 days postinoculation (Figure 2), representing a reasonable follow-up interval for therapeutic trials. Previous reports of successful engraftment of mucinous appendiceal tumours in mice did not document the time course of tumour growth in engrafted animals, stating only that ‘significant’ abdominal distension was noted within 1–3 months (Flatmark et al. 2007). We have demonstrated that intraperitoneal inoculation with fresh PMP-754 tumour suspension creates a relatively uniform population of engrafted mice for subsequent experimentation. Furthermore, we have experimentally documented that changes in abdominal girth and weight in tumour inoculated mice can be used as an indirect measure of tumour burden in reference to controls. The 100%‘take’ rate and time course of tumour growth are important observations for planning therapeutic intervention trials which were not experimentally validated in previous models (Flatmark et al. 2007).

Although we demonstrate that tumour can be frozen for up to 6 months and re-implanted in mice, the ‘take’ rate after re-implantation of frozen tumour is considerably lower than we typically encounter when transferring fresh tumour tissue. It would be advisable, therefore, to only include animals which have been inoculated with fresh tumour in therapeutic trials.

In our experience, mucinous appendiceal tumours are difficult to engraft and low-grade mucinous tumours resisted engraftment under our experimental conditions. A combination of technical issues and the biology of this type of tumour are the most likely explanations for this difficulty in initially establishing xenografts. We considered the possibility that our initial 8-week waiting period may be insufficient to obtain grossly recognizable intraperitoneal tumour growth in many cases. Recent experience suggests that a much longer waiting period (up to 8 months) may be necessary to establish initial xenografts (D. Bartlett, personal communication). Prolonged incubation may enable the engraftment of low-grade tumours.

We have demonstrated that the cytokeratin, mucin core protein and CDX2 immunophenotype (Figure 1 and Table 1) of PMP-754 is entirely consistent with other tumours of appendiceal origin (Yajima et al. 2005; Nonaka et al. 2006; Yoon et al. 2009) and is nearly identical to the resected primary tumour (Figure 1, Table 1).

Our studies also demonstrate that the primary tumour and PMP-754 exhibits aberrant nuclear expression of p53 and an elevated Ki-67 proliferative index (Figure 1, Table 1). Both elevated Ki-67 and p53 nuclear expression have been observed in appendiceal adenocarcinomas (Yajima et al. 2005; Yoon et al. 2009). These reports indicate that this expression may distinguish adenocarcinoma from low-grade appendiceal mucinous tumours (Yajima et al. 2005) or mucinous adenomas (Yoon et al. 2009).

Mutations in exons 5–8 of the TP53 gene have been documented in appendiceal adenocarcinomas with aberrant nuclear expression of p53 protein (Yajima et al. 2005). The heterogeneous pattern of p53 nuclear expression we observed is less likely to be associated with protein-stabilizing mutations in the TP53 gene than uniform, intense nuclear staining as has been demonstrated in several studies which have correlated immunohistochemical expression patterns with mutational analysis of TP53 (Esrig et al. 1993; Baas et al. 1994; Clausen et al. 1998). We evaluated the chromosomal region near the TP53 gene for LOH at 17p13 (D17S1289) and found no evidence of LOH in this region. Formal mutational analysis will be required to rule out the presence of a TP53 mutation, as LOH is not always detected in cases with a mutated allele (Ohue et al. 1994).

LOH at chromosome 10p23 (D10S1173) was detected in the xenograft tumour samples (Figure 3) but not in the original tumour. LOH was assessed in macrodissected tumour tissue which includes both stromal cells and neoplastic tumour epithelium. Appendiceal mucinous adenocarcinomas and low grade mucinous neoplasms of the appendix have a relatively low tumour epithelium to stromal cell ratio which makes purification of tumour epithelial cell DNA difficult. A high tumour DNA to normal DNA ratio facilitates detection of chromosomal copy number variations and other tumour-specific mutations. Enrichment of tumour epithelium occurs when pancreatic adenocarcinoma xenografts (Hahn et al. 1995). In primary resection specimens, pancreatic adenocarcinoma exhibits a low ratio of tumour epithelium to stroma due to the usually exuberant desmoplastic stromal response elicited by invasive pancreatic adenocarcinoma. Xenograft enrichment may explain the detection of LOH in PMP-754 but not resected tumour tissue. We cannot exclude the possibility, however, that a subclone of the resected tumour with LOH at 10p23 engrafted. It is unlikely that a de novo mutation occurred in PMP-754 during passage, given that LOH was analysed after only two passages and xenograft tumours are believed to be genetically stable (McQueen et al. 1991; Hahn et al. 1995).

Little is known about the molecular pathogenesis of appendiceal adenocarcinoma. This is undoubtedly due in large part to the relative rarity of these tumours. However, appendiceal neoplasms present unique challenges due to their paucicellularity. Engraftment of PMP-754 in nude mice enabled us to grow virtually unlimited quantities of fresh tumour for study and facilitated the molecular analysis we report herein. Based on our results, xenograft tumour ‘enrichment’ offers a plausible way to address our current lack of understanding of the molecular pathogenesis of this fatal disease.

To conclude, we have successfully xenografted an appendiceal mucinous adenocarcinoma. The tumour has been successfully transmitted to seven subsequent generations, it transfers with 100% reliability and tumour growth can be indirectly measured by following animal weight or abdominal girth. This well-characterized model will be useful to evaluate novel therapeutic strategies and methods of drug delivery for superficial, metastatic mucinous adenocarcinomas of the peritoneal cavity. Xenograft tumour enrichment has been used to facilitate the study of other tumours with low ratio of neoplastic tumour epithelium to stroma and holds considerable promise for the study of mucinous epithelial neoplasms of the appendix. Different strategies for engrafting low-grade mucinous appendiceal tumours, beginning with longer incubation times for initial intraperitoneal tumour implants, are needed to fully realize these experimental benefits and model the complete spectrum of appendiceal mucinous neoplasia.

References

  1. Baas IO, Mulder JW, Offerhaus GJ, Vogelstein B, Hamilton SR. An evaluation of six antibodies for immunohistochemistry of mutant p53 gene product in archival colorectal neoplasms. J. Pathol. 1994;172:5–12. doi: 10.1002/path.1711720104. [DOI] [PubMed] [Google Scholar]
  2. Baratti D, Kusamura S, Nonaka D, et al. Pseudomyxoma peritonei: clinical pathological and biological prognostic factors in patients treated with cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (HIPEC) Ann. Surg. Oncol. 2008;15:526–534. doi: 10.1245/s10434-007-9691-2. [DOI] [PubMed] [Google Scholar]
  3. Bradley RF, Stewart JHt, Russell GB, Levine EA, Geisinger KR. Pseudomyxoma peritonei of appendiceal origin: a clinicopathologic analysis of 101 patients uniformly treated at a single institution, with literature review. Am. J. Surg. Pathol. 2006;30:551–559. doi: 10.1097/01.pas.0000202039.74837.7d. [DOI] [PubMed] [Google Scholar]
  4. Carr NJ, Sobin LH. Unusual tumors of the appendix and pseudomyxoma peritonei. Semin. Diagn. Pathol. 1996;13:314–325. [PubMed] [Google Scholar]
  5. Clausen OP, Lothe RA, Borresen-Dale AL, et al. Association of p53 accumulation with TP53 mutations, loss of heterozygosity at 17p13, and DNA ploidy status in 273 colorectal carcinomas. Diagn. Mol. Pathol. 1998;7:215–223. doi: 10.1097/00019606-199808000-00006. [DOI] [PubMed] [Google Scholar]
  6. Esrig D, Spruck CH, III, Nichols PW, et al. p53 nuclear protein accumulation correlates with mutations in the p53 gene, tumor grade, and stage in bladder cancer. Am. J. Pathol. 1993;143:1389–1397. [PMC free article] [PubMed] [Google Scholar]
  7. Flatmark K, Reed W, Halvorsen T, et al. Pseudomyxoma peritonei – two novel orthotopic mouse models portray the PMCA-I histopathologic subtype. BMC Cancer. 2007;7:116. doi: 10.1186/1471-2407-7-116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Hahn SA, Seymour AB, Hoque AT, et al. Allelotype of pancreatic adenocarcinoma using xenograft enrichment. Cancer Res. 1995;55:4670–4675. [PubMed] [Google Scholar]
  9. Marsh JW, Finkelstein SD, Demetris AJ, et al. Genotyping of hepatocellular carcinoma in liver transplant recipients adds predictive power for determining recurrence-free survival. Liver Transpl. 2003;9:664–671. doi: 10.1053/jlts.2003.50144. [DOI] [PubMed] [Google Scholar]
  10. McQueen HA, Wyllie AH, Piris J, Foster E, Bird CC. Stability of critical genetic lesions in human colorectal carcinoma xenografts. Br. J. Cancer. 1991;63:94–96. doi: 10.1038/bjc.1991.19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Miner TJ, Shia J, Jaques DP, Klimstra DS, Brennan MF, Coit DG. Long-term survival following treatment of pseudomyxoma peritonei: an analysis of surgical therapy. Ann. Surg. 2005;241:300–308. doi: 10.1097/01.sla.0000152015.76731.1f. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Misdraji J, Yantiss RK, Graeme-Cook FM, Balis UJ, Young RH. Appendiceal mucinous neoplasms: a clinicopathologic analysis of 107 cases. Am. J. Surg. Pathol. 2003;27:1089–1103. doi: 10.1097/00000478-200308000-00006. [DOI] [PubMed] [Google Scholar]
  13. Mukherjee A, Parvaiz A, Cecil TD, Moran BJ. Pseudomyxoma peritonei usually originates from the appendix: a review of the evidence. Eur. J. Gynaecol. Oncol. 2004;25:411–414. [PubMed] [Google Scholar]
  14. Nonaka D, Kusamura S, Baratti D, Casali P, Younan R, Deraco M. CDX-2 expression in pseudomyxoma peritonei: a clinicopathological study of 42 cases. Histopathology. 2006;49:381–387. doi: 10.1111/j.1365-2559.2006.02512.x. [DOI] [PubMed] [Google Scholar]
  15. Ohue M, Tomita N, Monden T, et al. A frequent alteration of p53 gene in carcinoma in adenoma of colon. Cancer Res. 1994;54:4798–4804. [PubMed] [Google Scholar]
  16. Pai RK, Beck AH, Norton JA, Longacre TA. Appendiceal mucinous neoplasms: clinicopathologic study of 116 cases with analysis of factors predicting recurrence. Am. J. Surg. Pathol. 2009;33:1425–1439. doi: 10.1097/PAS.0b013e3181af6067. [DOI] [PubMed] [Google Scholar]
  17. Ronnett BM, Kurman RJ, Zahn CM, et al. Pseudomyxoma peritonei in women: a clinicopathologic analysis of 30 cases with emphasis on site of origin, prognosis, and relationship to ovarian mucinous tumors of low malignant potential. Hum. Pathol. 1995a;26:509–524. doi: 10.1016/0046-8177(95)90247-3. [DOI] [PubMed] [Google Scholar]
  18. Ronnett BM, Zahn CM, Kurman RJ, Kass ME, Sugarbaker PH, Shmookler BM. Disseminated peritoneal adenomucinosis and peritoneal mucinous carcinomatosis. A clinicopathologic analysis of 109 cases with emphasis on distinguishing pathologic features, site of origin, prognosis, and relationship to “pseudomyxoma peritonei”. Am. J. Surg. Pathol. 1995b;19:1390–1408. doi: 10.1097/00000478-199512000-00006. [DOI] [PubMed] [Google Scholar]
  19. Ronnett BM, Shmookler BM, Diener-West M, Sugarbaker PH, Kurman RJ. Immunohistochemical evidence supporting the appendiceal origin of pseudomyxoma peritonei in women. Int. J. Gynecol. Pathol. 1997;16:1–9. doi: 10.1097/00004347-199701000-00001. [DOI] [PubMed] [Google Scholar]
  20. Ronnett BM, Yan H, Kurman RJ, Shmookler BM, Wu L, Sugarbaker PH. Patients with pseudomyxoma peritonei associated with disseminated peritoneal adenomucinosis have a significantly more favorable prognosis than patients with peritoneal mucinous carcinomatosis. Cancer. 2001;92:85–91. doi: 10.1002/1097-0142(20010701)92:1<85::aid-cncr1295>3.0.co;2-r. [DOI] [PubMed] [Google Scholar]
  21. Smeenk RM, Verwaal VJ, Antonini N, Zoetmulder FA. Survival analysis of pseudomyxoma peritonei patients treated by cytoreductive surgery and hyperthermic intraperitoneal chemotherapy. Ann. Surg. 2007;245:104–109. doi: 10.1097/01.sla.0000231705.40081.1a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Smeenk RM, Bruin SC, van Velthuysen ML, Verwaal VJ. Pseudomyxoma peritonei. Curr. Probl. Surg. 2008;45:527–575. doi: 10.1067/j.cpsurg.2008.04.003. [DOI] [PubMed] [Google Scholar]
  23. Stewart JHT, Shen P, Russell GB, et al. Appendiceal neoplasms with peritoneal dissemination: outcomes after cytoreductive surgery and intraperitoneal hyperthermic chemotherapy. Ann. Surg. Oncol. 2006;13:624–634. doi: 10.1007/s10434-006-9708-2. [DOI] [PubMed] [Google Scholar]
  24. Sugarbaker PH. The natural history, gross pathology, and histopathology of appendiceal epithelial neoplasms. Eur. J. Surg. Oncol. 2006;32:644–647. doi: 10.1016/j.ejso.2006.03.016. [DOI] [PubMed] [Google Scholar]
  25. Sugarbaker PH, Jablonski KA. Prognostic features of 51 colorectal and 130 appendiceal cancer patients with peritoneal carcinomatosis treated by cytoreductive surgery and intraperitoneal chemotherapy. Ann. Surg. 1995;221:124–132. doi: 10.1097/00000658-199502000-00002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Szych C, Staebler A, Connolly DC, Wu R, Cho KR, Ronnett BM. Molecular genetic evidence supporting the clonality and appendiceal origin of Pseudomyxoma peritonei in women. Am. J. Pathol. 1999;154:1849–1855. doi: 10.1016/S0002-9440(10)65442-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Yajima N, Wada R, Yamagishi S, Mizukami H, Itabashi C, Yagihashi S. Immunohistochemical expressions of cytokeratins, mucin core proteins, p53, and neuroendocrine cell markers in epithelial neoplasm of appendix. Hum. Pathol. 2005;36:1217–1225. doi: 10.1016/j.humpath.2005.08.022. [DOI] [PubMed] [Google Scholar]
  28. Yoon SO, Kim BH, Lee HS, et al. Differential protein immunoexpression profiles in appendiceal mucinous neoplasms: a special reference to classification and predictive factors. Mod. Pathol. 2009;22:1102–1112. doi: 10.1038/modpathol.2009.74. [DOI] [PubMed] [Google Scholar]
  29. Young RH, Gilks CB, Scully RE. Mucinous tumors of the appendix associated with mucinous tumors of the ovary and pseudomyxoma peritonei. A clinicopathological analysis of 22 cases supporting an origin in the appendix. Am. J. Surg. Pathol. 1991;15:415–429. doi: 10.1097/00000478-199105000-00001. [DOI] [PubMed] [Google Scholar]
  30. Yousem SA, Nikiforova M, Nikiforov Y. The histopathology of BRAF-V600E-mutated lung adenocarcinoma. Am. J. Surg. Pathol. 2008;32:1317–1321. doi: 10.1097/PAS.0b013e31816597ca. [DOI] [PubMed] [Google Scholar]

Articles from International Journal of Experimental Pathology are provided here courtesy of Wiley

RESOURCES