Purpose: Among invasive ductal carcinomas of the pancreas (IDCP), there is a morphologically characteristic subgroup accompanied by abundant intraductal carcinoma components (ICCs). With the aim of determining whether ICC-rich IDCP are biologically different from ICC-poor IDCP, the expression status of Dpc4 protein was analyzed.

Experimental Design: A total of 43 IDCP was subdivided into two groups: (a) ICC-rich IDCP (ICCs area occupies ≥10% of the entire tumorous area); and (b) ICC-poor IDCP (with <10% of ICCs area). A total of 10 invasive carcinomas derived from intraductal papillary-mucinous neoplasms (ICs from IPMNs) were also analyzed. Each invasive and intraductal carcinoma area was then evaluated for Dpc4 protein status by immunohistochemistry.

Results: In a total of 43 IDCP, there were 23 ICC-rich IDCP and 20 ICC-poor IDCP. Dpc4-positive immunostaining was observed in the invasive carcinoma component of ICC-rich IDCP, ICC-poor IDCP, and ICs from IPMN in 18 of 23 (78%), 4 of 20 (20%), and 7 of 10 (70%) cases, respectively. In the intraductal component, positive staining for Dpc4 was found in 20 of 23 (87%), 3 of 7 (41%), and 8 of 10 (80%) cases, respectively. Dpc4 expression was found in both the invasive and ICC components of ICC-rich IDCP, similar to that found in IC derived from IPMN, whereas the expression of Dpc4 was largely diminished in ICC-poor IDCP.

Conclusions: Morphologically distinct subgroups of invasive ductal carcinomas of the pancreas, namely ICC-rich IDCP and ICC-poor IDCP, are also biologically distinguishable as revealed by the differential expression of Dpc4.

The invasive ductal carcinoma of the pancreas (IDCP) is a lethal disease, and the 5-year survival rate after the curative surgical resection ranges from 8 to 24% (1, 2). By careful histological analysis, some authors reported that among IDCP, about half of the cases were accompanied by intraductal proliferative lesions, which are classified under the pancreatic intraepithelial neoplasm (PanIN) classification (3, 4, 5). Many morphological and genetical investigations revealed that those PanINs were precursor lesions of IDCP (1, 4, 5, 6, 7, 8, 9, 10, 11, 12). We previously reported that among IDCP, about half of them were accompanied by abundant intraductal carcinoma components, corresponding to PanIN-3 of the PanIN classification (containing ≥10% than intraductal carcinoma area among the entire tumorous area; i.e., ICC rich) and were expected to have a more favorable prognosis than the other tumors that were poor in accompanying ICCs (containing <10% intraductal carcinoma area among the entire tumorous area; i.e., ICC poor; Ref. 3). Fukushima et al. (13) also reported that a longer survival was expected for patients whose IDCP had ICCs compared with those without.

Many genetic alterations, including K-ras, p53, p16, p21 and DPC4, have been investigated in IDCP (8, 12, 14). DPC4 is a tumor suppressor gene at chromosomal regions 18q21, of which, inactivation is one of the common genetic alterations identified in IDCP (6, 7, 8, 9, 11, 12, 14, 15, 16, 17, 18, 19, 20). In the cytoplasm, Dpc4 protein mediates signals from a family of tumor growth factor β ligands. The activation of the DPC4/tumor growth factor β signal pathway shows tumor suppressive functions as a result of inducing apoptosis or growth arrest in the G1 phase of the cell cycle (15). The immunohistochemical labeling for the Dpc4 protein is reported to be a sensitive and specific marker for DPC4 gene status, and there have been several genetic analyses of DPC4 in IDCP using immunohistochemical labeling as a genetic marker (19, 21). It was previously reported that Dpc4 protein was expressed in approximately half of all cases of IDCP (7, 17, 19, 21).

It is recognized that the intraductal papillary-mucinous neoplasms (IPMNs) of the pancreas are distinct tumors that have morphologically characteristic features and a more favorable prognosis than pancreatic IDCP (22, 23, 24, 25, 26). IPMNs consist of intraductal components with various histopathologically atypia from adenoma to carcinoma in situ, and they are sometimes associated with invasive carcinomas (e.g., invasive carcinoma derived from IPMN; IC from IPMN). Intraductal components accompanied by IC from IPMN were genetically revealed to be the precursor lesions of invasive carcinoma components (27). It was previously reported that Dpc4 protein was expressed in virtually all IPMNs in contrast to the situation in IDCP (20, 23, 28).

In this study, Dpc4 protein status was analyzed in morphologically distinct subgroups of pancreatic cancer, i.e., ICC-rich IDCP, ICC-poor IDCP, and IC from IPMN, with the aim of determining what, if any, biological differences exist between the different subtypes. Surprisingly, Dpc4 protein expression was also found in the invasive component of ICC-rich IDCP, as was the case with IC from IPMN.

Patients and Specimens.

A total of 61 IDCP of the pancreas was surgically resected at the National Cancer Center Hospital East from 1992 to 2000. Cases excluded from the study comprised 4 patients who underwent palliative resection and 10 patients who had unusual adenocarcinomas (5 cases were accompanied by adenosquamous carcinoma components and 5 were accompanied by anaplastic carcinoma components). Four patients who died within 1 month of surgery because of postoperative complications were also excluded. Finally, a total of 43 cases was enrolled in this study: 21 men and 22 women. The mean age was 60.3 years, ranging from 25 to 80 years. The mean tumor size was 36.5 mm, ranging from 17 to 110 mm, and tumors were located in the pancreatic head in 32 cases, the pancreatic body in 8, and the pancreatic tail in 3. Thirty-two patients underwent pancreaticoduodenectomies, including 31 pylorus-preserving pancreaticoduodenectomies and 1 conventional Whipple resection, and 11 underwent distal pancreatectomies. Thirty-two patients received intraoperative radiation therapy. All resected specimens were fixed in 10% formalin overnight at room temperature and cut stepwise at 5- to 7-mm intervals. Specimens were processed routinely and embedded in paraffin. From each paraffin block, serial sections were cut and stained with H&E. All tumors were evaluated according to the guidelines of the WHO (26) and the pathological Tumor-Node-Metastasis classification (29). Fourteen cases were classified as well-differentiated adenocarcinoma, 21 as moderately differentiated adenocarcinoma, and 8 as poorly differentiated adenocarcinoma. Ten cases had 0 or 1 positive lymph nodes and in 33 2 or more lymph nodes were positive. Two cases were classified as stage II under the pathological Tumor-Node-Metastasis classification, 22 as stage III, and 19 as stage IV. There was no case with distant organ metastasis. Surgical margin statuses were negative in 33 cases and positive in 10. Duration of follow-up was 66.3 months, ranging from 27 to 115 months. Three cases are currently alive and 40 have died (Table 1).

A total of 10 ICs from IPMN was also examined, which were resected at the National Cancer Center Hospital East from 1992 to 2001: 7 men and 3 women. The mean tumor size was 77 mm, ranging from 40 to 122 mm. Because the number of ICs from IPMN was too small, we could not perform an analysis of the outcomes of the patients with ICs from IPMN.

Evaluation of Intraductal Carcinoma Components.

We previously reported that IDCP containing ≥10% intraductal carcinoma area among the entire tumorous area, defined as ICC rich IDCP, were expected to be associated with a significantly more favorable prognosis than those containing <10% intraductal carcinoma area among the entire tumorous area, defined as ICC-poor IDCP (3). One section of each block was stained with Elastica to confirm noninvasive tumor components, and we determined as ICCs only those cancerous lesions representing papillary, low papillary, or cribriforming encircled by elastic fibers (Fig. 1, A and B). We calculated the proportion of ICCs and classified all tumors into ICC-rich or ICC-poor IDCP; 23 cases were classified as ICC-rich IDCP and 20 as ICC-poor IDCP. There was no significant difference in any of the clinicopathological characteristics between the two groups (Table 1).

Immunohistochemical Study.

We screened H&E-stained slides of each case by light microscopy and selected two to four paraffin blocks from each ICC-rich IDCP and IC from IPMN, containing normal pancreatic tissue, an invasive carcinoma area, and an intraductal carcinoma area. We also selected two to three paraffin blocks from each ICC-poor IDCP, containing normal pancreatic tissue, an invasive carcinoma area, and an intraductal carcinoma area if available. For Dpc4 immunohistochemical labeling, 3-μm unstained serial sections were cut from the selected paraffin blocks, and each section was deparaffined in xylene and rehydrated in alcohol. Endogenous peroxidase activity was blocked in 3% hydrogen peroxide in methanol. Antigen retrieval was achieved by microwave heating at 95°C for 20 min in citrate buffer at pH 7.4. After cooling in room air for 60 min, each section was incubated with a 1:100 dilution of monoclonal antibody to Dpc4 protein (clone B8; Santa Cruz Biotechnology, Santa Cruz, CA) for 12 h at 4°C. Nonspecific bindings of the secondary antibody were blocked by immersing in 2% normal swine serum PBS, and the biotinylated secondary antibody, avidin-biotin complex, and 3,3′-diaminobenzidine chromagens were applied to detect the anti-Dpc4 antibody. Sections were counterstained with hematoxylin and Elastica. The normal pancreatic tissue in each section served as an internal positive control, and the primary antibody was omitted in the negative control. Two of the authors (H. Takahashi and T. Hasebe) assessed the histological findings and immunohistochemical staining of the carcinomas. Whenever there was a disagreement between the two authors, they reexamined the section and reached a consensus through discussion. In each case, the invasive carcinoma area and the intraductal carcinoma area were evaluated separately. The immunohistochemical staining of Dpc4 protein was scored as positive when >50% of neoplastic cells were stained as strongly as the surrounding normal pancreas and the other cases as negative (28, 30).

Laser Capture Microdissection and Reverse Transcription-PCR.

To confirm the specificity of immunohistochemistry with Dpc4, reverse transcription-PCR was performed on glands staining positive and negative for Dpc4 protein using material obtained by laser capture microdissection system (PixCell-II; Arcturus Engineering, Santa Clara, CA). In brief, the dehydrated 10-μm frozen tissue section was overlaid with a thermoplastic membrane mounted on optically transparent caps, and Dpc4-positive and -negative glands (each corresponding to 500–1000 cells) were captured by focal melting of the membrane through laser activation. The captured tissues were then immersed in denaturation Trizol solution (Life Technologies, Inc., Gaithersburg, MD), and total RNA was extracted. RNA was redissolved in 10 μl of oligo(dT)20 primer solution, and cDNAs were synthesized using the ThermoScript RT-PCR System (Life Technologies, Inc.) in a final volume of 20 μl. Quantitative real-time PCR was carried out using a LightCycler instrument (Roche, Mannheim, Germany). One μl of cDNA solution was subjected to 40 PCR cycles of 10 s at 95°C, 10 s at 53°C–65°C, and 5–15 s at 72°C in a 10-μl mixture containing 1 μl of LightCycler-DNA Master SYBR Green I (Roche), 2.25–5 mm MgCl2, and 0.25 μm each of forward (5′-TCCCTGGCCCAGGATCAGTA-3′) and reverse (5′-CTGATAAGGTTAAGGGCCCCAAC-3′) DPC4-specific primers. An external standard curve for the DPC4 gene was generated using serial 102-fold dilutions of reverse transcription-PCR products, corresponding to 1 × 108 to 1 × 102 copies/μl. Because the amount of obtained cells and subsequent RNA in Dpc4-positive and -negative glands were different, mRNA copy number of DPC4 was standardized by dividing by the copy number of glyceraldehyde-3-phosphate dehydrogenase.

Statistical Analysis.

The following clinicopathological factors were compared by the proportion of intraductal carcinoma area and Dpc4 protein status: (a) gender; (b) age; (c) tumor size; (d) tumor location; (e) lymph node status; (f) the degree of differentiation of tumor; (g) pathological Tumor-Node-Metastasis staging; (h) surgical margin status; and (i) use of intraoperative radiation therapy. Fisher’s exact probability test was used to assess cross-tabulations, and Mann-Whitney’s U test was used to compare the mean values. Univariate survival analyses on each parameter were performed by a log-rank test, and the survival curves were drawn by the Kaplan-Meier method. The parameters that showed a significant correlation with the prognosis in the univariate analyses were then entered into the multivariate analysis using the Cox proportional hazards regression model (31). All tests were two tailed, and P < 0.05 was accepted as statistically significant. Statistical analyses were performed using Statview 5.0 software (Abacus Systems, Berkeley, CA).

Immunohistochemical Study.

All normal cells of pancreas such as islet cells, acinar cells, fibroblasts, and nonneoplastic ductal cells expressed Dpc4 protein. The Dpc4 protein status in the invasive carcinoma area of 43 IDCP was positive in 22 cases (51%) and negative in 21 (49%; Fig. 2, A and B). ICCs were observed in 30 of 43 IDCP, and their Dpc4 protein status was positive in 23 cases (77%) and negative in 7 (23%) (Fig. 1, A and B). Forty-three IDCP were subdivided into 23 ICC-rich IDCP and 20 ICC-poor IDCP. The Dpc4 protein status in the intraductal and invasive carcinoma areas of IC-rich IDCP was positive in 20 (87%) and 18 (78%), respectively. Among 20 ICC-poor IDCP, 7 were accompanied by ICC, and the Dpc4 protein status in their intraductal and invasive carcinoma areas was positive in 3 (41%) and 4 (20%), respectively (Table 2). The invasive carcinoma areas of ICC-rich IDCP showed a significantly higher frequency of Dpc4 protein expression than those of ICC-poor IDCP, and a significantly higher frequency of Dpc4 protein expression in the intraductal carcinoma areas was also observed in ICC rich IDCP than in ICC poor IDCP (Fig. 3).

In 10 ICs from IPMN, the Dpc4 protein status in invasive carcinoma area and intraductal carcinoma was positive in 7 cases (70%) and 8 (80%), respectively (Table 2). The Dpc4 protein status in ICC rich IDCP was very similar to that in ICs from IPMN in both invasive carcinoma and intraductal carcinoma areas (Fig. 3).

Expression of mRNAs for DPC4 in Immunohistochemically Positive and Negative Glands.

Each pancreatic cancer glands positively and negatively stained for Dpc4 were obtained by laser capture microdissection. Representative pictures before (Fig. 4,A) and after (Fig. 4,B) removal and the isolated cell adherent to the membrane (Fig. 4,C) were shown. Expression of DPC4 mRNA copies/105 copies glyceraldehyde-3-phosphate dehydrogenase was 316.8 and 10.1 for Dpc4 staining-positive and -negative glands, respectively, confirming the validity of immunohistochemistry as a genetic marker for DPC4 inactivation (Fig. 4 D).

Correlations between Clinicopathological Characteristics and the Dpc4 Protein Status in IDCP Patients.

The Dpc4 protein status was significantly associated only with age in the univariate analyses (Table 3), and patients with Dpc4 protein-positive IDCP were significantly younger than those who were Dpc4 protein-negative IDCP (P < 0.001).

Univariate and Multivariate Analysis for Prognosis.

In the univariate analysis, the proportion of intraductal carcinoma area, ICC rich or ICC poor, was a statistically significant factor for improved prognosis in the univariate analysis (P < 0.05), as we reported previously (Ref. 3; Fig. 5). The median survival times of ICC-rich IDCP and ICC-poor IDCP patients were 15.0 and 10.0 months, respectively. Also, the Dpc4 protein status was a statistically significant factor for an improved prognosis (P < 0.05; Fig. 6). The median survival times for patients whose tumors expressed Dpc4 protein and those who did not were 15.5 and 10.0 months, respectively. In addition, the following factors were statistically associated with a poor prognosis of IDCP patients in the univariate analyses: (a) tumor size of <35 versus ≥35 mm (P < 0.05); and (b) surgical margin status (P < 0.01). The Cox proportional hazards regression model using the above significant factors in the univariate analyses showed that the proportion of ICCs and tumor size was found to be independent significant factors for improving outcomes of patients with IDCP (Table 4).

It is generally accepted that DPC4 expression is frequently diminished in invasive pancreatic carcinomas. However, our data indicate that a subgroup of IDCP exists in which the expression of DPC4 is intact, which corresponds to the morphologically distinct subgroup having an abundant component of intraductal carcinoma.

There have been several studies that have immunohistochemically evaluated the Dpc4 protein status of IDCP and IPMNs. They showed that the frequency of DPC4 inactivation in the invasive carcinoma and intraductal carcinoma areas in IDCP ranged from 49 to 56% and from 31 to 41%, respectively (9, 17, 18, 19, 21). In IPMN, the rates of Dpc4 protein expression in the invasive and intraductal carcinoma areas ranged from 80 to 97% and from 91 to 100%, respectively (20, 23, 28). In this study, the Dpc4 protein status in IDCP was positive for 22 of 43 (51%) in the invasive carcinoma area and 23 of 30 (77%) in the intraductal carcinoma area and that in IC from IPMN was positive for 7 of 10 (70%) in the invasive carcinoma area and 8 of 10 (80%) in the intraductal carcinoma area, respectively. These results are in agreement with those reported previously.

It was previously reported that the presence of ICCs in IDCP was a significant good prognostic parameter for patients with IDCP after surgical resection and the survival time of patients with ICC-rich IDCP was almost similar to that of those with ICs from IPMNs (3, 13, 22, 23, 24, 25). Furthermore, some studies referred to the morphological similarities of intraductal carcinoma components between IDCP accompanied with ICCs and IPMN and the difficulties in distinguishing between ICC-rich IDCP and ICs from IPMNs (3, 4, 13, 25). This study clearly showed that the frequencies of Dpc4 protein expression of ICC-rich IDCP were significantly higher than that of ICC-poor IDCP in invasive and intraductal carcinoma area and were similar to that in ICs from IPMN. These results suggested that ICC-rich IDCP were different from ICC-poor IDCP and rather similar to ICs from IPMNs, morphologically, clinically, biologically, and possibly in carcinogenesis. To verify this hypothesis, genetic alterations that will be able to distinguish IDCP and ICs from IPMN are required to investigate and perform a detailed comparative study among ICC-rich IDCP, ICC-poor IDCP, and IPMNs.

In the analyses of the prognosis, only the tumor size (<35 or ≥35 mm) and the proportion of intraductal carcinoma area were independently prognostic. Tascilar et al. (21) reported that the Dpc4 protein status in IDCP was an independent prognostic factor in a multivariate analysis of patients with IDCP. However, in our study, the prognostic predictive power of the Dpc4 protein status was inferior to the tumor size or the proportion of intraductal carcinoma area. Because a close relationship between the proportion of intraductal carcinoma area and the Dpc4 protein status was observed in this study, the former probably has a stronger prognostic power than the latter. Thus, we conclude that the proportion of intraductal carcinoma area in IDCP is the better prognostic parameter than the Dpc4 protein status for patients with IDCP of the pancreas.

In summary, Dpc4 is expressed in ICC-rich IDCP, similar to ICs from IPMN. In ICC-poor IDCP, Dpc4 expression is largely absent. The present results suggest that two different pathways exist for pancreatic carcinogenesis. The first is a classical multistep PanIN pathway in which DPC4 inactivation plays a critical role in the late stages of carcinogenesis and is represented by ICC-poor IDCP in present study (corresponding to 20 of 43, 47% of cases). The second carcinogenic pathway might be termed the IPMN Pathway, in which DPC4 is not involved in malignancy. These were represented by ICC-rich IDCP that accounted for 53% (23 of 43) of IDCP and had similar biological and morphological characteristics to ICs from IPMN. The relationship between morphological presentation and Dpc4 status, however, remains to be elucidated. Although there may be cases that convert from ICC-rich IDCP to ICC-poor IDCP, classifying conventional bulk IDCP into those two biologically, morphologically, and clinically different subgroups may be useful in further understanding pancreatic carcinogenesis and developing new treatment strategies for these cancers.

Grant support: A Grant-in-Aid for Cancer Research from the Ministry of Health, Labor and Welfare and a Grant-in-Aid for the Second Term Comprehensive 10-Year Strategy for Cancer Control from the Ministry of Health, Labor and Welfare in Japan.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Requests for reprints: Atsushi Ochiai, National Cancer Center Hospital East, 6-5-1, Kashiwanoha, Kashiwa, Chiba 277-8577, Japan. Phone: 81-04-7133-1111; Fax: 81-04-7131-4724; E-mail: aochiai@east.ncc.go.jp

Fig. 1.

Intraductal carcinoma component in invasive ductal carcinoma. Tumor cells extending in low papillary features are observed (A). Elastic fibers of pancreatic ducts are stained blue, and intraductal carcinoma components with low papillary features show a positive cytoplasmic staining for Dpc4 protein (B). Magnification, ×100.

Fig. 1.

Intraductal carcinoma component in invasive ductal carcinoma. Tumor cells extending in low papillary features are observed (A). Elastic fibers of pancreatic ducts are stained blue, and intraductal carcinoma components with low papillary features show a positive cytoplasmic staining for Dpc4 protein (B). Magnification, ×100.

Close modal
Fig. 2.

Dpc4 immunohistochemical labeling in the invasive ductal carcinomas (IDCs). IDC of pancreas expressing Dpc4 protein (A), compared with invasive ductal carcinoma of pancreas that does not express Dpc4 protein (B). All normal cells of pancreas around the tumor cells express Dpc4 protein. Magnification, ×200.

Fig. 2.

Dpc4 immunohistochemical labeling in the invasive ductal carcinomas (IDCs). IDC of pancreas expressing Dpc4 protein (A), compared with invasive ductal carcinoma of pancreas that does not express Dpc4 protein (B). All normal cells of pancreas around the tumor cells express Dpc4 protein. Magnification, ×200.

Close modal
Fig. 3.

Dpc4 protein expression rate of ICC rich invasive ductal carcinomas (IDCs), intraductal carcinoma component (ICC)-poor IDCs and ICs from intraductal papillary-mucinous neoplasm (IPMN). The Dpc4 protein status of ICC-rich IDCs was significantly higher than that of ICC-poor IDCs in the intraductal and invasive carcinoma areas (∗ P < 0.05; ∗∗ P < 0.001, calculated using Fisher’s exact probability test). The Dpc4 protein status of ICs from IPMN was as high as that of ICC-rich IDCs in the intraductal and invasive carcinoma areas.

Fig. 3.

Dpc4 protein expression rate of ICC rich invasive ductal carcinomas (IDCs), intraductal carcinoma component (ICC)-poor IDCs and ICs from intraductal papillary-mucinous neoplasm (IPMN). The Dpc4 protein status of ICC-rich IDCs was significantly higher than that of ICC-poor IDCs in the intraductal and invasive carcinoma areas (∗ P < 0.05; ∗∗ P < 0.001, calculated using Fisher’s exact probability test). The Dpc4 protein status of ICs from IPMN was as high as that of ICC-rich IDCs in the intraductal and invasive carcinoma areas.

Close modal
Fig. 4.

Quantitative reverse transcription-PCR of laser-microdissected Dpc4-positive and -negative glands. To confirm the specificity of immunohistochemistry with Dpc4, reverse transcription-PCR was performed on areas staining positive and negative for Dpc4 protein using material obtained by laser capture microdissection. Representative pictures before (A) and after (B) microdissection and the isolated cell adherent to the membrane (C) were shown. Expression of DPC4 mRNA copies in staining positive and negative glands by quantitative real-time PCR analysis using a LightCycler. The amount of DPC4 mRNA in positive glands was higher than that in negative glands (D), confirming the validity of immunohistochemistry as a genetic marker for DPC4 inactivation.

Fig. 4.

Quantitative reverse transcription-PCR of laser-microdissected Dpc4-positive and -negative glands. To confirm the specificity of immunohistochemistry with Dpc4, reverse transcription-PCR was performed on areas staining positive and negative for Dpc4 protein using material obtained by laser capture microdissection. Representative pictures before (A) and after (B) microdissection and the isolated cell adherent to the membrane (C) were shown. Expression of DPC4 mRNA copies in staining positive and negative glands by quantitative real-time PCR analysis using a LightCycler. The amount of DPC4 mRNA in positive glands was higher than that in negative glands (D), confirming the validity of immunohistochemistry as a genetic marker for DPC4 inactivation.

Close modal
Fig. 5.

Survival after operation of invasive ductal carcinoma (IDC) stratified for the proportion of intraductal carcinoma area, intraductal carcinoma component (ICC)-rich or ICC poor. ICC-rich IDC patients show a significantly longer survival period than ICC-poor IDC patients (P < 0.05).

Fig. 5.

Survival after operation of invasive ductal carcinoma (IDC) stratified for the proportion of intraductal carcinoma area, intraductal carcinoma component (ICC)-rich or ICC poor. ICC-rich IDC patients show a significantly longer survival period than ICC-poor IDC patients (P < 0.05).

Close modal
Fig. 6.

Survival after operation of invasive ductal carcinoma stratified for the SMAD4 status. Invasive ductal carcinoma patients with SMAD4 protein expression show a significantly longer survival period than those without SMAD4 protein expression (P < 0.05).

Fig. 6.

Survival after operation of invasive ductal carcinoma stratified for the SMAD4 status. Invasive ductal carcinoma patients with SMAD4 protein expression show a significantly longer survival period than those without SMAD4 protein expression (P < 0.05).

Close modal
Table 1

Clinicopathological characteristics of 43 invasive ductal carcinomas of pancreas stratified for ICCsa proportion

Characteristicsn (%)ICCsP
RichbPoorc
23 (53)20 (47)
Gender    n.s.d 
 Male 21 (49) 11 (48) 10 (50)  
 Female 22 (51) 12 (52) 10 (50)  
Age (years, mean ± SD) 60.3 ± 12.5 59.2 ± 12.5 61.5 ± 12.7 n.s.e 
Tumor size (mm, mean ± SD) 36.5 ± 16.0 35.5 ± 12.5 37.6 ± 19.4 n.s.e 
 <35 mm 20 (47) 10 (43) 10 (50) n.s.d 
 ≥35 mm 23 (53) 13 (57) 10 (50)  
Tumor location    n.s.d 
 Ph 32 (74) 17 (74) 15 (75)  
 Pb 8 (19)    
 Pt 3 (7) 6 (26) 5 (25)  
Nodes positive    n.s.d 
 0 or 1 10 (23) 6 (26) 4 (20)  
 >1 33 (77) 17 (74) 16 (80)  
Differentiation    n.s.d 
 Well differentiated 14 (33) 16 (70) 19 (95)  
 Moderately differentiated 21 (49)    
 Poorly differentiated 8 (19) 7 (30) 1 (5)  
Pathological Tumor-Node-Metastasis stage    n.s.d 
 I 0 (0) 14 (61) 10 (50)  
 II 2 (5)    
 III 22 (51)    
 IV 19 (44) 9 (39) 10 (50)  
Surgical treatment    n.s.d 
 PD 32 (74) 18 (78) 14 (70)  
 DP 11 (26) 5 (22) 6 (30)  
Surgical margin status    n.s.d 
 Negative 33 (77) 17 (74) 16 (80)  
 Positive 10 (23) 6 (26) 4 (20)  
Intraoperative radiation therapy    n.s.d 
 Yes 32 (74) 16 (70) 16 (80)  
 No 11 (26) 7 (30) 4 (20)  
Vital status    n.s.d 
 Alive 3 (7) 3 (13) 0 (0)  
 Dead 40 (93) 20 (87) 20 (100)  
Duration of follow-up (months, mean ± SD) 66.3 ± 26.0 64.3 ± 23.5 68.6 ± 29.1 n.s.e 
Characteristicsn (%)ICCsP
RichbPoorc
23 (53)20 (47)
Gender    n.s.d 
 Male 21 (49) 11 (48) 10 (50)  
 Female 22 (51) 12 (52) 10 (50)  
Age (years, mean ± SD) 60.3 ± 12.5 59.2 ± 12.5 61.5 ± 12.7 n.s.e 
Tumor size (mm, mean ± SD) 36.5 ± 16.0 35.5 ± 12.5 37.6 ± 19.4 n.s.e 
 <35 mm 20 (47) 10 (43) 10 (50) n.s.d 
 ≥35 mm 23 (53) 13 (57) 10 (50)  
Tumor location    n.s.d 
 Ph 32 (74) 17 (74) 15 (75)  
 Pb 8 (19)    
 Pt 3 (7) 6 (26) 5 (25)  
Nodes positive    n.s.d 
 0 or 1 10 (23) 6 (26) 4 (20)  
 >1 33 (77) 17 (74) 16 (80)  
Differentiation    n.s.d 
 Well differentiated 14 (33) 16 (70) 19 (95)  
 Moderately differentiated 21 (49)    
 Poorly differentiated 8 (19) 7 (30) 1 (5)  
Pathological Tumor-Node-Metastasis stage    n.s.d 
 I 0 (0) 14 (61) 10 (50)  
 II 2 (5)    
 III 22 (51)    
 IV 19 (44) 9 (39) 10 (50)  
Surgical treatment    n.s.d 
 PD 32 (74) 18 (78) 14 (70)  
 DP 11 (26) 5 (22) 6 (30)  
Surgical margin status    n.s.d 
 Negative 33 (77) 17 (74) 16 (80)  
 Positive 10 (23) 6 (26) 4 (20)  
Intraoperative radiation therapy    n.s.d 
 Yes 32 (74) 16 (70) 16 (80)  
 No 11 (26) 7 (30) 4 (20)  
Vital status    n.s.d 
 Alive 3 (7) 3 (13) 0 (0)  
 Dead 40 (93) 20 (87) 20 (100)  
Duration of follow-up (months, mean ± SD) 66.3 ± 26.0 64.3 ± 23.5 68.6 ± 29.1 n.s.e 
a

ICC, intraductal carcinoma component; n.s., not significant; PD, pancreaticoduodenectomy; DP, distal pancreatectomy.

b

ICCs rich means that ICCs area occupies ≥10% of entire tumorous area.

c

ICCs poor means that ICCs area occupies <10% of entire tumorous area.

d

P was calculated using Fisher’s exact probability test.

e

P was calculated using Mann-Whitney’s U test.

Table 2

Dpc4 status of ICCa-rich IDC, ICC-poor IDC, and IC from IPMN

TumorsDpc4 protein expression (%)
Intraductal carcinoma areaInvasive carcinoma area
CasesPositiveNegativeCasesPositiveNegative
ICC poor IDCPb 7c 3 (41) 4 (59)d 20 4 (20) 16 (80)e 
ICC rich IDCPf 23 20 (87) 3 (13)d 23 18 (78) 5 (22)e 
IC from IPMN 10 8 (80) 2 (20) 10 7 (70) 30 (30) 
TumorsDpc4 protein expression (%)
Intraductal carcinoma areaInvasive carcinoma area
CasesPositiveNegativeCasesPositiveNegative
ICC poor IDCPb 7c 3 (41) 4 (59)d 20 4 (20) 16 (80)e 
ICC rich IDCPf 23 20 (87) 3 (13)d 23 18 (78) 5 (22)e 
IC from IPMN 10 8 (80) 2 (20) 10 7 (70) 30 (30) 
a

ICC, intraductal carcinoma component; IDC, invasive ductal carcinoma; IPMN, intraductal papillary-mucinous neoplasm; IDCP, IDC pancreas.

b

ICCs poor means that IC area occupies <10% of entire tumorous area.

c

Only 7 cases out of 20 ICC-poor IDCP accompanied by ICC area, and no ICC area was found in the rest 13 cases.

d

P < 0.05, calculated using Fisher’s exact probability test.

e

P < 0.001, calculated using Fisher’s exact probability test.

f

ICCs rich means that IC area occupies ≥10% of entire tumorous area.

Table 3

Clinicopathological characteristics of IDCsa stratified for Dpc4 protein status

CharacteristicsDpc4 protein status in invasive carcinoma areaP
Positive (%)Negative (%)
22 (51)21 (49)
Gender   n.s.b 
 Male 10 (45) 11 (52)  
 Female 12 (55) 10 (48)  
Age (years, mean ± SD) 56.3 ± 12.0 64.4 ± 11.8 <0.01c 
Tumor size (mm, mean ± SD) 34.8 ± 12.6 38.2 ± 19.0 n.s.c 
 <35 mm 10 (45) 10 (48) n.s.b 
 ≥35 mm 12 (55) 11 (52)  
Tumor location   n.s.b 
 Ph 15 (68) 17 (81)  
 Pbt 7 (32) 4 (19)  
Nodes positive   n.s.b 
 >1 18 (82) 15 (71)  
 0 or 1 4 (18) 6 (29)  
Differentiation   n.s.b 
 Well differentiated 10 (45) 4 (19)  
 Moderate/poorly differentiated 12 (55) 17 (81)  
Pathological Tumor-Node-Metastasis stage classification   n.s.b 
 II/III 12 (55) 12 (57)  
 IV 10 (45) 9 (43)  
Surgical margin status   n.s.b 
 Positive 4 (18) 6 (29)  
 Negative 18 (82) 15 (71)  
Intraoperative radiation therapy   n.s.b 
 Yes 16 (73) 16 (76)  
 No 6 (27) 5 (24)  
CharacteristicsDpc4 protein status in invasive carcinoma areaP
Positive (%)Negative (%)
22 (51)21 (49)
Gender   n.s.b 
 Male 10 (45) 11 (52)  
 Female 12 (55) 10 (48)  
Age (years, mean ± SD) 56.3 ± 12.0 64.4 ± 11.8 <0.01c 
Tumor size (mm, mean ± SD) 34.8 ± 12.6 38.2 ± 19.0 n.s.c 
 <35 mm 10 (45) 10 (48) n.s.b 
 ≥35 mm 12 (55) 11 (52)  
Tumor location   n.s.b 
 Ph 15 (68) 17 (81)  
 Pbt 7 (32) 4 (19)  
Nodes positive   n.s.b 
 >1 18 (82) 15 (71)  
 0 or 1 4 (18) 6 (29)  
Differentiation   n.s.b 
 Well differentiated 10 (45) 4 (19)  
 Moderate/poorly differentiated 12 (55) 17 (81)  
Pathological Tumor-Node-Metastasis stage classification   n.s.b 
 II/III 12 (55) 12 (57)  
 IV 10 (45) 9 (43)  
Surgical margin status   n.s.b 
 Positive 4 (18) 6 (29)  
 Negative 18 (82) 15 (71)  
Intraoperative radiation therapy   n.s.b 
 Yes 16 (73) 16 (76)  
 No 6 (27) 5 (24)  
a

IDC, invasive ductal carcinoma; n.s., not significant; PD, pancreaticoduodenectomy; DP, distal pancreatectomy.

b

P was calculated using Fisher’s exact probability test.

c

P was calculated using Mann-Whitney’s U test.

Table 4

Multivariate analysis for prognosis of patients with IDCs

CharacteristicsTotal (n = 43)Mortality cases (%)Hazard ratio95% confidence intervalP
ICCsa      
 Richb 23 20 (87) Referent   
 Poorc 20 20 (100) 2.62 1.28–5.36 0.008 
Tumor size      
 <35 mm 20 17 (85) Referent   
 ≥35 mm 23 23 (100) 2.49 1.27–4.90 0.008 
CharacteristicsTotal (n = 43)Mortality cases (%)Hazard ratio95% confidence intervalP
ICCsa      
 Richb 23 20 (87) Referent   
 Poorc 20 20 (100) 2.62 1.28–5.36 0.008 
Tumor size      
 <35 mm 20 17 (85) Referent   
 ≥35 mm 23 23 (100) 2.49 1.27–4.90 0.008 
a

ICC, intraductal carcinoma component; IDC, invasive ductal carcinoma.

b

ICCs rich means that IDC area occupies ≥10% of entire tumorous area.

c

ICCs poor means that IDC area occupies <10% of entire tumorous area.

1
Conlon KC, Klimstra DS, Brennan MF. Long-term survival after curative resection for pancreatic ductal adenocarcinoma. Clinicopathologic analysis of 5-year survivors.
Ann Surg
,
223
:
273
-9,  
1996
.
2
Yeo CJ, Cameron JL. Prognostic factors in ductal pancreatic cancer.
Langenbecks Arch Surg
,
383
:
129
-33,  
1998
.
3
Kawahira H, Hasebe T, Kinoshita T, et al The intraductal carcinoma component is a significant prognostic parameter in patients with invasive ductal carcinoma of the pancreas.
Jpn J Cancer Res
,
93
:
1138
-44,  
2002
.
4
Hruban RH, Adsay NV, Albores-Saavedra J, et al Pancreatic intraepithelial neoplasia: a new nomenclature and classification system for pancreatic duct lesions.
Am J Surg Pathol
,
25
:
579
-86,  
2001
.
5
Cubilla AL, Fitzgerald PJ. Morphological lesions associated with human primary invasive nonendocrine pancreas cancer.
Cancer Res
,
36
:
2690
-8,  
1976
.
6
Brat DJ, Lillemoe KD, Yeo CJ, Warfield PB, Hruban RH. Progression of pancreatic intraductal neoplasias to infiltrating adenocarcinoma of the pancreas.
Am J Surg Pathol
,
22
:
163
-9,  
1998
.
7
Heinmoller E, Dietmaier W, Zirngibl H, et al Molecular analysis of microdissected tumors and preneoplastic intraductal lesions in pancreatic carcinoma.
Am J Pathol
,
157
:
83
-92,  
2000
.
8
Hruban RH, Goggins M, Parsons J, Kern SE. Progression model for pancreatic cancer.
Clin Cancer Res
,
6
:
2969
-72,  
2000
.
9
Wilentz RE, Iacobuzio-Donahue CA, Argani P, et al Loss of expression of Dpc4 in pancreatic intraepithelial neoplasia: evidence that DPC4 inactivation occurs late in neoplastic progression.
Cancer Res
,
60
:
2002
-6,  
2000
.
10
Klein WM, Hruban RH, Klein-Szanto AJ, Wilentz RE. Direct correlation between proliferative activity and dysplasia in pancreatic intraepithelial neoplasia (PanIN): additional evidence for a recently proposed model of progression.
Mod Pathol
,
15
:
441
-7,  
2002
.
11
Yamano M, Fujii H, Takagaki T, Kadowaki N, Watanabe H, Shirai T. Genetic progression and divergence in pancreatic carcinoma.
Am J Pathol
,
156
:
2123
-33,  
2000
.
12
Hruban RH, Wilentz RE, Kern SE. Genetic progression in the pancreatic ducts.
Am J Pathol
,
156
:
1821
-5,  
2000
.
13
Fukushima N, Sakamoto M, Mukai K, et al Intraductal papillary components in invasive ductal carcinoma of the pancreas are associated with long-term survival of patients.
Hum Pathol
,
32
:
834
-41,  
2001
.
14
Gerdes B, Ramaswamy A, Ziegler A, et al p16INK4a is a prognostic marker in resected ductal pancreatic cancer: an analysis of p16INK4a, p53, MDM2, an Rb.
Ann Surg
,
235
:
51
-9,  
2002
.
15
Liu F. SMAD4/DPC4 and pancreatic cancer survival. Commentary re: M. Tascilar et al., The Dpc4 protein and prognosis of pancreatic ductal adenocarcinoma, Clin Cancer Res., 7: 4115–4121, 2001.
Clin Cancer Res
,
7
:
3853
-6,  
2001
.
16
Hahn SA, Schutte M, Hoque AT, et al DPC4, a candidate tumor suppressor gene at human chromosome 18q21.1.
Science (Wash. DC)
,
271
:
350
-3,  
1996
.
17
Hruban RH, Offerhaus GJ, Kern SE, Goggins M, Wilentz RE, Yeo CJ. Tumor-suppressor genes in pancreatic cancer.
J Hepatobiliary Pancreat Surg
,
5
:
383
-91,  
1998
.
18
Luttges J, Galehdari H, Brocker V, et al Allelic loss is often the first hit in the biallelic inactivation of the p53 and DPC4 genes during pancreatic carcinogenesis.
Am J Pathol
,
158
:
1677
-83,  
2001
.
19
Wilentz RE, Su GH, Dai JL, et al Immunohistochemical labeling for dpc4 mirrors genetic status in pancreatic adenocarcinomas: a new marker of DPC4 inactivation.
Am J Pathol
,
156
:
37
-43,  
2000
.
20
Luttges J, Zamboni G, Longnecker D, Klöppel G. The immunohistochemical mucin expression pattern distinguishes different types of intraductal papillary mucinous neoplasms of the pancreas and determines their relationship to mucinous noncystic carcinoma and ductal adenocarcinoma.
Am J Surg Pathol
,
25
:
942
-8,  
2001
.
21
Tascilar M, Skinner HG, Rosty C, et al The Dpc4 protein and prognosis of pancreatic ductal adenocarcinoma.
Clin Cancer Res
,
7
:
4115
-21,  
2001
.
22
Yamao K, Ohashi K, Nakamura T, et al The prognosis of intraductal papillary mucinous tumors of the pancreas.
Hepatogastroenterology
,
47
:
1129
-34,  
2000
.
23
Sohn TA, Yeo CJ, Cameron JL, Iacobuzio-Donahue CA, Hruban RH, Lillemoe KD. Intraductal papillary mucinous neoplasms of the pancreas: an increasingly recognized clinicopathologic entity.
Ann Surg
,
234
:
313
-21,  
2001
.
24
Adsay NV, Conlon KC, Zee SY, Brennan MF, Klimstra DS. Intraductal papillary-mucinous neoplasms of the pancreas: an analysis of in situ and invasive carcinomas in 28 patients.
Cancer (Phila.)
,
94
:
62
-77,  
2002
.
25
Loftus EV, Olivares-Pakzad BA, Batts KP, et al Intraductal papillary-mucinous tumors of the pancreas: clinicopathologic features, outcome, and nomenclature. Members of the Pancreas Clinic, and Pancreatic Surgeons of Mayo Clinic.
Gastroenterology
,
110
:
1909
-18,  
1996
.
26
Klöppel G, Solcia E, Longnecker DS, Capella C, Sobin LH. .
Histologic typing of tumors of the exocrine pancreas
, 2nd ed. Springer-Verlag Geneva  
1996
.
27
Fujii H, Inagaki M, Kasai S, et al Genetic progression and heterogeneity in intraductal papillary-mucinous neoplasms of the pancreas.
Am J Pathol
,
151
:
1447
-54,  
1997
.
28
Iacobuzio-Donahue CA, Klimstra DS, Adsay NV, et al Dpc-4 protein is expressed in virtually all human intraductal papillary mucinous neoplasms of the pancreas: comparison with conventional ductal adenocarcinomas.
Am J Pathol
,
157
:
755
-61,  
2000
.
29
Sobin LH, Wittekind CH. .
UICC TNM Classification of Malignant Tumors
, 5th ed. Springer-Verlag Berlin  
1997
.
30
Biankin AV, Biankin SA, Kench JG, et al Aberrant p16(INK4A) and DPC4/Smad4 expression in intraductal papillary mucinous tumours of the pancreas is associated with invasive ductal adenocarcinoma.
Gut
,
50
:
861
-8,  
2002
.
31
Cox DR. Regression models and life-tables.
J R Stat Soc
,
34
:
187
-220,  
1972
.