Annals of Surgical Oncology Cite Track
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
10.1245/s10434-006-9321-4
Annals of Surgical Oncology 14:1694-1702 (2007)
© 2007 Society of Surgical Oncology
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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 Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mori, K.
Right arrow Articles by Sasaki, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mori, K.
Right arrow Articles by Sasaki, H.

Original Article

Detection of Minimal Gastric Cancer Cells in Peritoneal Washings by Focused Microarray Analysis with Multiple Markers: Clinical Implications

Kazuhiko Mori, MD1,6, Tomohiro Suzuki, PhD3, Hiroshi Uozaki, MD2, Hayao Nakanishi, MD4, Tetsuya Ueda, MD1, Yoshihiro Matsuno, MD7, Yasuhiro Kodera, MD5, Hiromi Sakamoto, PhD6, Nobuko Yamamoto, PhD3, Mitsuru Sasako, MD8, Michio Kaminishi, MD1 and Hiroki Sasaki, PhD6

1 Department of Gastrointestinal Surgery, University of Tokyo Graduate School of Medicine, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-8655, Japan
2 Department of Pathology, University of Tokyo Graduate School of Medicine, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-0033, Japan
3 Nanotechnology and Advanced System Research Laboratories, Canon Inc., Shimomaruko 3-30-2, Ohta-ku, Tokyo 146-8501, Japan
4 Laboratory of Pathology, Aichi Cancer Center Research Institute, Aichi Cancer Center Hospital, Kanokoden 1-1, Chikusa-ku, Nagoya, Aichi 464-8681, Japan
5 Department of Gastroenterological Surgery, Nagoya University Graduate School of Medicine, 65, Tsurumai-cho, Showa-ku, Nagoya, Aichi 466-8550, Japan
6 Genetics Division, National Cancer Center Research Institute, National Cancer Center, Tsukiji 5-1-1, Chuo-ku, Tokyo 104-0045, Japan
7 Clinical Laboratory, National Cancer Center Hospital, Tsukiji 5-1-1, Chuo-ku, Tokyo 104-0045, Japan
8 Surgical Oncology Department, National Cancer Center Hospital, Tsukiji 5-1-1, Chuo-ku, Tokyo 104-0045, Japan

Correspondence: Address correspondence and reprint requests to: Hiroki Sasaki, PhD; E-mail: hksasaki{at}gan2.res.ncc.go.jp


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Peritoneal cytology is an important prognostic factor of gastric cancer. However, peritoneal cytology requires great skill, which may explain its low prevalence. A reverse transcriptase–polymerase chain reaction–based assay with multiple marker genes or immunocytochemistry was assessed as an alternative method of gathering the same kind of data as cytology.

Methods: Peritoneal washings from 179 patients with gastric cancer were analyzed by multiplex reverse transcriptase–polymerase chain reaction with 10 marker genes and subsequent hybridization to a customized oligo-nucleotide array. Results with this assay were either validated as a prognostic factor or confirmed by demonstrating the presence of cancer cells by immunocytochemical cytology.

Results: Only 1 (2.2%) of 44 disease-free cases was shown to be positive by the microarray assay, whereas 13 (93%) of 14 conventional cytology–positive cases were found to be positive. This assay further detected approximately one-third of cytology-negative patients either with peritoneal recurrence (7 of 20, 35%) or with non-peritoneal recurrence (6 of 22, 27%). A high concordance between the microarray assay and immunocytochemical cytology with five antibodies against CK20, FABP1, MUC2, TFF1, and MASPIN was confirmed. The clinical outcome of the microarray assay–positive cases was poor, as was that of the cytology-positive cases.

Conclusions: Our assay, though time-consuming and requiring special equipment, demonstrated a specificity and sensitivity equal to or better than cytology in our institutes. The minimal free peritoneal cancer cells detected by the microarray assay may provide the same clinical information as larger amounts of cancer cells for patients with gastric cancer. An anti-MASPIN antibody may be helpful in peritoneal cytology of gastric cancer.

Key Words: Gastric cancer • Peritoneal cytology • RT-PCR • Microarray • MASPIN


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Gastric cancer is one of the most frequently encountered malignancies.1 Although it can often be cured through surgical resection, leaving no evident residual disease, recurrences are experienced after a potentially curative surgical operation, with peritoneal metastasis as the most common form.2 Therefore, peritoneal cytology as an assessment of the risk for peritoneal recurrence is often performed for advanced disease, and it is recognized as one of the most important prognostic factors for gastric cancer.3 Beginning in 1999, the Japanese Gastric Cancer Association began including a description of the intraoperative peritoneal wash cytology as "CY status" in its documentation system, and any surgical procedure labeled as CY1 (cytology positive) is considered noncurative.4 In addition, usage of laparoscopic disease evaluation and peritoneal cytology before surgical resection is being advocated for patients with advanced gastric cancer.5 Neoadjuvant therapies may be indicated as a potentially better treatment option for cases with T3 or T4 disease (tumor penetrates serosa or invades adjacent structures) or positive peritoneal cytology proven by a laparoscopic staging system.6 Thus, peritoneal cytology is establishing an increasing clinical implication in the management of advanced gastric cancer.

However, because conventional peritoneal cytology provides only a limited sensitivity, short-time peritoneal recurrences or even synchronous peritoneal metastases are not rare in cytology negative cases. Specimens containing few atypical cells or containing indeterminate cells mimicking reactive mesothelium are often defined as negative, perhaps because the job requires great skill by trained cytologists (or pathologists). With these problems, the peritoneal cytology in gastric cancer operation is not prevailing in spite of its importance. Trained cytologists are always used in large specialist hospitals, such as those of a university or cancer center, but are often unavailable at other general hospitals. Our goals were twofold: first, to discover a molecular biological method that could be used in a clinical laboratory agency as a substitute for skilled cytology; and second, to ensure that the procedure has uniform accuracy, regardless of the site where the procedure is performed.

To improve the sensitivity of peritoneal cytology, reverse transcriptase–polymerase chain reaction (RT-PCR)-based methods or immunocytochemical cytology have been introduced.7–9 However, these challenges experienced problems resulting from the use of a single marker gene or only a few marker genes to detect cancer cells even though the cancer cells show expressional variations from cell to cell and from case to case. For example, CEA has been reported to be lacking in specificity and CK20 in sensitivity.10,11 We previously identified 11 genes as potentially useful markers: CK20, FABP1, MUC2, TFF1, TFF2, MASPIN, GW112, PRSS4, MDK, SOX9, and CDX1. Among them, five markers, CK20, FABP1, MUC2, TFF1, and TFF2, showed highly specific results in nested RT-PCR for peritoneal cytology of gastric cancer.12 Considering the advantage in the use of multiple markers,13 we here report a sensitive microarray assay that uses multiple marker genes for detecting minimal cancer cells in peritoneal washings of patients with gastric cancer. Concurrently, we also validated the usefulness of our previously identified five marker genes (CK20, FABP1, MUC2, TFF1, and MASPIN) in immunocytochemical peritoneal cytology.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Clinical Materials
A total of 179 peritoneal wash samples from patients with gastric cancer who gave written informed consent were collected at the National Cancer Center (NCC, 98 cases), the Aichi Cancer Center (ACC, 47 cases), and the University of Tokyo Hospital (34 cases). The NCC samples were those used in our previous studies,12 and the ACC samples were representative, and thus not consecutive, cases (18 stage Ia cases, 12 recurrence-free cases, and 17 relapsed cases) described in another previous study.7 Each set of samples derived from these three institutes was used for three analyses: the capability of predicting recurrence, the impact on clinical outcome, and concordance with findings of immunostaining of the assay. Our study design was approved in the institutional review board of each institute. The method for the RNA extraction is detailed in our previous reports.7,10

Marker Gene Set
After identifying 11 potentially useful marker genes (CK20, FABP1, MUC2, TFF1, TFF2, MA-SPIN, GW112, PRSS4, MDK, SOX9, and CDX1) in our previous study,12 we further screened additional thousands of genes for marker genes with a new version of microarray GeneChip U133A (Affimetrix, Santa Clara, CA) by use of the same gene screening procedures.12 We first selected four marker candidates, PRSS2, PRSS8, TACSTD1, and BCAS1. However, 6 (MDK, SOX9, CDX1, PRSS2, PRSS8, and BCAS1) of the 15 genes showed clear false-positive signals in representative clinical samples with an earlier version of a focused microarray containing the 15 genes and a most popular marker CEA. Thus, these six genes were excluded. Finally, we selected 10 markers, CK20, FABP1, MUC2, TFF1, TFF2, MASPIN, GW112, PRSS4, TAC-STD1, and CEA.

Manufacturing Microarrays for Detecting Minimal Gastric Cancer Cells in Peritoneal Washings
Focused microarray is constructed by fixing 50–60 mer of oligonucleotide probes on a slide glass by Bubble Jet technology.14 The microarray contains a single spot for each sequence of 10 marker genes and ß-actin as a control gene. Each probe sequence used for the microarray is listed in Table 1Go.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Sequences of primers and probes
 
Marker Gene Detection by Focused Microarray
As shown in a schematic flow diagram (Fig. 1Go), the MiniChip assay consists of four steps: (1) amino allyl–dUTP labeling by multiplex RT-PCR; (2) coupling the labeled RT-PCR products to Cy3 dye; (3) hybridization Cy3-labeled cDNA to microarray; and (4) fluorescence scanning. Representative microarray data of cytology-positive and -negative cases are shown in Fig. 2Go. This assay was able to detect fluorescence signals in cases where RT-PCR products could not be visualized even by a highly sensitive capillary electrophoresis system such as the Agilent Bioanalyzer 2100 (Fig. 2Go).


Figure 1
View larger version (54K):
[in this window]
[in a new window]

  		FIG. 1. Schematic flow diagram of MiniChip assay. Microscopic photo depicts one peritoneal wash cytology slide stained by anti-MASPIN antibody. Atypical cells are rare (arrows). Marker transcripts among extracted total RNAs were amplified and labeled with amino allyl–dUTP by multiplex reverse transcriptase–polymerase chain reaction (step 1) and stained by coupling to Cy3 dye (step 2), followed by hybridization to focused microarray (step 3) and fluorescence intensity scanning (step 4).

 

Figure 2
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 2. Image of focused microarrays. and distribution of Mini-Chip assay results. Positions of probes for marker genes (left, top). Images of microarrays (left, bottom) and electrophoreses (right), respectively, show results obtained by MiniChip assay and capillary electrophoresis system from three representative cases. (a) Stage Ia case. (b) Cytology-negative case with peritoneal recurrence. (c) Cytology-positive case. L, marker ladder.

 
Multiplex RT-PCR and Coupling of Cy3 Dye
From .5 to 1 µg of total RNA prepared from 20–50 mL of peritoneal washings, reverse transcription was performed with Superscript II (Invitrogen, Carlsbad, CA) with random hexamer in a total volume of 20 µL according to the manufacturer’s protocol. Multiplex RT-PCR was performed in two tubes at different PCR cycles: 30 cycles for relatively cancer-specific genes, CK20, FABP1, MUC2, TFF1, TFF2, and CEA, and 25 cycles for MASPIN, GW112, PRSS4, and TACSTD1 together with ß-actin. PCR primer sequences are listed in Table 1Go. Ten microliters of the PCR solution in each tube consisted of 1 µL of template cDNA, primers (2.5 pmol each), 50 µM of amino allyl–dUTP, 1 µL of AccuPrime 10 x buffer 1 (2 mM of dNTP, 15 mM of MgCl2) and .25 µL of AccuPrime Taq polymerase (Invitrogen). A thermal cycler was set with initial heating at 94°C for 2 minutes, followed by an amplification cycle heated at 94°C for 15 seconds, 60°C for 45 seconds, and 72°C for 3 minutes. The two PCR solutions were mixed and purified with QIAquick PCR purification Kit (Qiagen, Tokyo, Japan). Purified cDNAs were dried by evaporation and dissolved in 9 µL of .2 M carbonate buffer (pH 9.0). Two microliters of Cy3 monoreactive dye (Amersham Biosciences, Piscataway, NJ) was added to the cDNA solution, followed by incubation at room temperature for 1 hour. After two cycles of this process, cDNA was purified by QiaQuick purification kit.

Hybridization to Focused Array and Fluorescence Scanning
The entire Cy3-labeled cDNA solution (50 µL) was mixed in 120 µL of a hybridization cocktail (6 x buffer containing 900 mM of NaCl, 60 mM of NaH2PO4, and H2O, and 6 mM of EDTA, pH 7.4/10% formamide/.05% sodium dodecyl sulfate). By a hybridization apparatus, HybStation (Genomic Solutions, Ann Arbor, MI), an array was preheated to 65°C for 3 minutes, filled with the hybridization cocktail, then incubated at 92°C for 2 minutes and then at 55°C for 4 hours. Subsequently, the array was washed with 2 x standard saline citrate (SSC), .1% sodium dodecyl sulfate at 25°C and then 2 x SSC at 20°C, and rinsed with .1 x SSC, in accordance with the steps laid out in a conventional manual, and finally dried in a spin drier. The array was scanned by an apparatus for DNA microarrays (Genepix 4000B; Axon Instruments, Union City, CA), and the fluorescence intensity from each probe spot was obtained after subtracting the background level. Fluorescence intensity from a ß-actin probe was used as an internal control so that the ratio of fluorescence intensity to that of ß-actin could be analyzed.

Establishing Diagnostic Criteria for MiniChip Assay
A putative cutoff value for each gene was established from the fluorescence ratio of 39 samples derived from stage Ia cases (cancer confined to the submucosa without nodal involvement) as negative control cases because cases of this stage seldom develop peritoneal metastasis.15 Two cutoff values were attempted, one at a value corresponding to the maximum value plus standard deviation (MAX SD) and the other at the average value plus twice standard deviation (AVG 2SD) of the 39 stage Ia cases. On the basis of the fluorescence ratios between each marker gene and ß-actin under these two cutoff values, we defined in this study any case with two or more positive markers as a MiniChip assay–positive case because we ascertained that tumor-negative samples may well be weakly positive by a single marker, whereas a tumor-negative sample with two or more positive markers would be rare.

Analyses of Clinical Outcome
The clinical outcome of the NCC and ACC patients was investigated to evaluate the sensitivity and specificity for the disease recurrence of the MiniChip assay. Patients of the two institutes were followed up in a same manner with clinical imagings and measurement of carcinoembryonic antigen (CEA) and cancer antigen (CA) 19-9 every 3 to 6 months.7,10 Cases with at least 2 years (700 days) of follow-up and without a recent history of other malignancies were eligible for the analyses of clinical outcome. For NCC patients, the disease-free survival was analyzed by Kaplan-Meier curve with a diagnosis of disease recurrence as an end point to compare the impact of the MiniChip assay and its clinical outcome with that of conventional cytology. ACC cases were selected representative cases and thus were not used for the survival analyses. Samples from the University of Tokyo Hospital lacked adequate clinical follow-up time and thus were used only for the immunocytochemistry below.

Immunocytochemistry
To evaluate the usefulness of our previously identified five marker genes (CK20, FABP1, MUC2, TFF1, and MASPIN) in immunocytochemical peritoneal cytology, we used anti-MASPIN (BD Pharmingen, San Diego, CA), anti-CK20 (Dako, Kyoto, Japan), anti-TFF1 (Dako), anti-MUC2 (Novo-Castra, New Castle, UK), and anti-FABP1 protein (Abcam, Cambridge, UK) as primary antibodies (all are mouse monoclonal IgG) and a Histofine Simple Stain Max PO (M) (Nichirei, Tokyo, Japan) as peroxidase-conjugated secondary antibody. By use of four gastric cell lines, the mRNA levels of the five genes of which have been previously examined,12 the quality of all primary antibody products was verified and an optimal dilution ratio of each antibody was established.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Application of the Focused Microarray Analysis to Detect Minimal Gastric Cancer Cells in Peritoneal Washings and Its Clinical Impact
In our previous report12 and in this study, we selected the 10 genes as tumor cell–specific genes. In fact, 6 genes (CK20, FABP1, MUC2, TFF1, TFF1, and CEA) of the 10 have never been detected in early cancers (tumor cell–negative cases) despite nested RT-PCR with outer and inner primer sets, and the other 4 genes (MASPIN, GW112, PRSS4, and TACSTD1) have also shown no or quite weak bands in the early cancers with high 30 PCR cycles (12 for MASPIN, GW112, and PRSS4, and data not shown in TACSTD1). Therefore, the MiniChip assay belongs in a negative or positive assay. However, it is required for determining the cutoff values.

The distribution of fluorescence ratios between each marker gene and ß-actin in 39 stage Ia–early cancer cases, 44 disease-free cases, and 65 metastatic cases are plotted in Fig. 3Go. High-level signals of five genes, TFF1, TFF2, CK20, FABP1, and TACSTD1, were found to be obviously more specific to the above metastatic cases. Therefore, as described in detail in Methods, we attempted two methods for the diagnosis: in one, all 10 genes were used in the diagnosis with a cutoff set at MAX SD, and in the other, 5 genes with a cutoff set at AVG 2SD were used.


Figure 3
View larger version (36K):
[in this window]
[in a new window]

  		FIG. 3. Distribution of MiniChip assay results. Contrast among three groups of cases is clear in microarray images, whereas electrophoresis system detects polymerase chain reaction products only in cytology-positive cases. Plots on left, middle, and right rows are fluorescence ratios of stage Ia, disease-free, and metastatic cases, respectively. Dashed line indicates level of average value plus twice standard deviation (AVG 2SD) of stage Ia cases; solid line indicates maximum value plus standard deviation (MAX SD). Upper five genes, TFF1, TFF2, FABP1, CK20, and TACSTD1, showed more specific results than other genes.

 
After excluding the 39 early cancer cases used in the establishment of the cutoff values and 18 cases with synchronous metastases, 62 cases (30 patients experienced relapse) at the NCC and 29 cases (17 patients experienced relapse) at the ACC were eligible as a validation set for predicting recurrence.

The number of the two kinds of MiniChip assay–positive cases at both the NCC and the ACC and the conventional cytology–positive cases in the 62 NCC cases and recurrence status are shown in Table 2Go. Comparing the two cutoff sets, the cutoff set at AVG 2SD for five genes showed better results in the diagnosis than that with a cutoff set at MAX SD for 10 genes. Therefore, we focused on the results with a cutoff set at AVG 2SD for five genes.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Number of cases defined as cancer positive by MiniChip assay and conventional cytology
 
Only 1 case (2.2%) of 44 disease-free cases (1 of 32 at the NCC and 0 of 12 at the ACC) was shown to be positive by the MiniChip assay, whereas 13 (93%) of 14 conventional cytology–positive cases (6 of 6 at the NCC and 7 of 8 at the University of Tokyo Hospital; Tables 2Go and 3Go) were found to be positive. These results demonstrate that this assay has low false-positive and false-negative findings. In 14 NCC cases with peritoneal recurrence, the MiniChip assay detected more than did the conventional cytology (eight cases vs. five), and accordingly, three of nine cytology-negative patients with peritoneal recurrence were detected by the MiniChip assay. For the 29 ACC cases, which included only cytology-negative patients, the MiniChip assay detected 4 of 11 cases with peritoneal recurrence. The MiniChip assay detected approximately one-third (7 of 20, 35%) of cases found to be falsely negative by conventional cytology at both institutes. Interestingly, the MiniChip assay also detected 3 of 16 NCC cases and 3 of 6 ACC cases with nonperitoneal recurrence (6 of 22, 27%), in accordance with our previous finding.12


View this table:
[in this window]
[in a new window]

  		TABLE 3. MiniChip assay and immunostaining
 
As shown in Fig. 4Go, the Kaplan-Meier curve demonstrates the impact of the MiniChip assay on the disease status of patients with gastric cancer treated with potentially curative surgery. The clinical outcome of the MiniChip assay–positive patients (MiniChip positive, CY0) as well as conventional cytology–positive patients (CY1) was found to be quite poor compared with the MiniChip assay–negative patients with negative conventional cytology (MiniChip negative, CY0) (log rank test, P < .001).


Figure 4
View larger version (8K):
[in this window]
[in a new window]

  		FIG. 4. MiniChip assay results and conventional cytology and relationship to disease-free survival. Cases defined as cancer positive by MiniChip assay but negative by conventional cytology (MiniChip positive, CY0) showed disease-free curve similar to that of cytology positive cases (CY1) and had significantly worse prognosis compared with both negative cases (MiniChip negative, CY0) (log rank test, P < .001).

 
High Concordance Between the MiniChip Assay and Immunocytochemical Cytology
To evaluate the usefulness of the five marker genes (CK20, FABP1, MUC2, TFF1, and MASPIN) in immunocytochemical cytology, we performed immunostaining with available antibodies for these genes’ products in 34 cases at the University of Tokyo Hospital. The results of the MiniChip assay, immunostaining, and conventional cytology accompanied by peritoneal metastatic status as an operative finding are summarized in Table 3Go. Of the 34 cases, 14 were found to be positive by MiniChip assay. These 14 cases included seven of eight cases found to be positive by conventional cytology. Of the 14 cases, anti-MASPIN, anti-CK20, aniti-TFF1, anti-MUC2, and anti-FABP1 detected 12 (86%), 9 (64%), 6 (43%), 4 (29%), and 4 (29%) cases, respectively. Two cases (patients 7 and 11) were not detected with any of the antibodies used. One case (patient 15), whose results were negative by MiniChip assay but positive by conventional cytology, was detected with four antibodies except anti-FABP1. All of the five cases with peritoneal metastasis confirmed as an operative finding were detected by both the MiniChip assay and immunostaining. These results suggest that immunocytochemical cytology with the five antibodies as well as the MiniChip assay could contribute to the prediction of cancer recurrence. Among the five antibodies, anti-MASPIN was highly specific to atypical cells and rarely stained noncancerous cells. Representative immunostaining results are shown in Fig. 5Go.


Figure 5
View larger version (132K):
[in this window]
[in a new window]

  		FIG. 5. Atypical cells found in conventional cytology–negative but MiniChip assay–positive cases by immunocytochemistry. Stained slides by anti-MASPIN and anti-CK20 antibodies in two cases. (1) Case 9, MASPIN (x 40). (2) Case 9, MASPIN (x 200). (3) Case 9 CK20 (x 200). (4) Case 13, MASPIN (x 40). (5) Case 13, MASPIN (x 200). (6) Case 13, CK20 (x 200). Immunocytochemistry clearly demonstrated atypical cells in conventional cytology–negative slides and confirmed positive findings of MiniChip assay.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Although the role of peritoneal cytology is established as an important tool in the management of gastric cancer, it is nevertheless a job that requires great skill by trained cytologists, which may account for its low prevalence in clinical practice. However, cytologists should be able to find support from the MiniChip assay and/or immunocytochemical analysis, in particular with anti-MASPIN antibody, demonstrated here to have an improved and stable sensitivity for a minimal amount of cancer cells. Present results of immunocytochemistry with anti-MASPIN antibody encourage its application in the detection of micrometastases or isolated tumor cells in the lymph nodes for which anti-CK20 was frequently used in previous studies.11,16

In the MiniChip assay, five genes, TFF1, TFF2, FABP1, CK20, and TACSTD1, showed highly specific results; however, CEA and MUC2 were unexpectedly less contributing, despite our previous findings.7,12 This is possibly because of inefficient or nonspecific amplification by multiplex RT-PCR. For evaluating the sensitivity of the MiniChip assay, a fraction (estimated at 3.8 x 103 cells) of gastric cancer cell line HSC60 was serially diluted (1:4n) by a peritoneal wash sample from one early gastric cancer patient and was analyzed by the MiniChip assay. Reproducible positive results with the MiniChip assay were observed from samples diluted as much as 1:16, suggesting that the threshold for the detection of the assay was approximately 200 cells per 1.5 mL of peritoneal washings (data not shown). Although the sensitivity of RT-PCR–based methods is expected to be higher than that of immunocytochemistry,17 both of the two methods, the MiniChip assay and immunocytochemistry, detected twice as many positive cases as did conventional cytology. Therefore, further optimization of the primers or probes, including those for additional markers recently identified,18,19 is required to improve the results.

As shown in Fig. 4Go, the disease-free survival curve of the MiniChip assay–positive patients was almost identical to that of the conventional cytology–positive patients. If this observation is consistent with a larger cohort of patients, rare free peritoneal tumor cells, beyond the sensitivity limit of conventional cytology, detected by the MiniChip assay should be interpreted as indicating a poor prognosis with recurrence likely to occur shortly—as short a time as that in cases with positive conventional cytology. However, because the benefits of adjuvant therapies in some solid tumors were reported to relate to the amount of remnant tumor burden after potentially curative surgery,20,21 patients with a MiniChip assay–positive result might show a better overall survival after adjuvant therapies compared with those with a positive conventional cytology result.

Our present data suggest an improved sensitivity and reliability of the MiniChip assay and immunocytochemical cytology by anti-MASPIN compared with conventional cytology. Additional information about the status of remnant tumor burden known through these assays might be helpful for future clinical trials to identify a cluster of patients who would most benefit from adjuvant therapies.


   ACKNOWLEDGMENTS
 
Supported in part by a grant from the program for the Promotion of Fundamental Studies in Health Sciences of the National Institute of Biochemical Innovation (NiBio); and by a Grant-in-Aid for the Third Comprehensive 10-Year Strategy for Cancer Control and for Cancer Research (15-5 and 16-15) from the Ministry of Health, Labour and Welfare of Japan. K.M. was a recipient of Research Resident Fellowships from the Foundation for Promotion of Cancer Research. We thank Mr. Kiyoaki Nomoto for his assistance in immunocytochemistry.

Received for publication August 17, 2006. Accepted for publication November 9, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Ferlay J, Bray F, Pisani P, et al. Globocan 2000: Cancer Incidence, Mortality, and Prevalence Worldwide. Ver 1.0, vol 5. Lyon: IARC Press, 2001.
  2. Wu CW, Lo SS, Shen KH, et al. Incidence and factors associated with recurrence patterns after intended curative surgery for gastric cancer. World J Surg 2003; 27:153–8.[Medline]
  3. Bonenkamp JJ, Songun I, Hermans J, et al. Prognostic value of positive cytology findings from abdominal washings in patients with gastric cancer. Br J Surg 1996; 83:672–4.[Medline]
  4. Japanese Gastric Cancer Association Japanese classification of gastric carcinoma. 2nd English ed. Gastric Cancer 1998; 1:10–24.[Medline]
  5. Onate-Ocana LF, Gallardo-Rincon D, Aiello-Crocifoglio V, et al. The role of pretherapeutic laparoscopy in the selection of treatment for patients with gastric carcinoma: a proposal for a laparoscopic staging system. Ann Surg Oncol 2001; 8:624–31.[Abstract/Free Full Text]
  6. Yano M, Yasuda T, Fujiwara Y, et al. Preoperative intra-peritoneal chemotherapy for patients with serosa-infiltrating gastric cancer. J Surg Oncol 2004; 88:39–43.[CrossRef][Medline]
  7. Kodera Y, Nakanishi H, Ito S, et al. Quantitative detection of disseminated free cancer cells in peritoneal washes with real-time reverse transcriptase–polymerase chain reaction. Ann Surg 2002; 235:499–506.[CrossRef][Medline]
  8. Fujimura T, Ohta T, Kitagawa H, et al. Trypsinogen expression and early detection for peritoneal dissemination in gastric cancer. J Surg Oncol 1998; 69:71–5.[CrossRef][Medline]
  9. Benevolo M, Mottolese M, Cosimelli M, et al. Diagnostic and prognostic value of peritoneal immunocytology in gastric cancer. J Clin Oncol 1998; 16:3406–11.[Abstract]
  10. Nagao K, Hisatomi H, Hirata H, et al. Expression of molecular marker genes in various types of normal tissue: implication for detection of micrometastases. Int J Mol Med 2002; 10:307–10.[Medline]
  11. Kubota K, Nakanishi H, Hiki N, et al. Quantitative detection of micrometastases in the lymph nodes of gastric cancer patients with real-time RT-PCR: a comparative study with immunohistochemistry. Int J Cancer 2003; 105:136–43.[CrossRef][Medline]
  12. Mori K, Aoyagi K, Ueda T, et al. Highly specific marker genes for detecting minimal gastric cancer cells in cytology negative peritoneal washings. Biochem Biophys Res Commun 2004; 313:931–7.[CrossRef][Medline]
  13. Taback B, Chan AD, Kuo CT, et al. Detection of occult metastatic breast cancer cells in blood by a multimolecular marker assay: correlation with clinical stage of disease. Cancer Res 2001; 61:8845–50.[Abstract/Free Full Text]
  14. Okamoto T, Suzuki T, Yamamoto N. Microarray fabricaton with covalent attachment of DNA using Bubble Jet technology. Nat Biotech 2000; 18:438–41.[CrossRef][Medline]
  15. Ushijima T, Sasako M. Focus on gastric cancer. Cancer Cell 2004; 5:121–5.[CrossRef][Medline]
  16. Ishigami S, Natsugoe S, Tokuda K, et al. Clinical impact of micrometastasis of the lymph node in gastric cancer. Am Surg 2003; 69:573–7.[Medline]
  17. Mattano LA Jr, Moss TJ, Emerson SG. Sensitive detection of rare circulating neuroblastoma cells by the reverse transcriptase-polymerase chain reaction. Cancer Res 1992; 52:4701–5.[Abstract/Free Full Text]
  18. Sakakura C, Takemura M, Hagiwara A, et al. Overexpression of dopa decarboxylase in peritoneal dissemination of gastric cancer and its potential as a novel marker for the detection of peritoneal micrometastases with real-time RT-PCR. Br J Cancer 2004; 90:665–71.[CrossRef][Medline]
  19. Wang ZN, Xu HM, Jiang L, et al. Expression of survivin in primary and metastatic gastric cancer cells obtained by laser capture microdissection. World J Gastroenterol 2004; 10:3094–8.[Medline]
  20. Harstrick A, Klaassen U, Eberhardt W, et al. Role of chemotherapy against micrometastatic cancer. Cancer Metastasis Rev 1999; 18:151–72.[CrossRef][Medline]
  21. Day RS, Shackney SE, Peters WP. The analysis of relapse-free survival curves: implications for evaluating intensive systemic adjuvant treatment regimens for breast cancer. Br J Cancer 2005; 92:47–54.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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 Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mori, K.
Right arrow Articles by Sasaki, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mori, K.
Right arrow Articles by Sasaki, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS