Recently, the effectiveness of preoperative chemotherapy for cancer treatment has attracted much attention. The authors conducted a study to investigate the influence of 5'-DFUR administration on the cell cycle, based on DNA analysis using FCM in patients with breast cancer. A total of 53 patients with breast cancer (Stages I, II and III) were investigated concerning tumor PyNPase activity and cell cycle phase. Subjects were stratified according to tumor diameter, histologic type and ER in the group receiving preoperative 5'-DFUR administration (25 patients) and the group not receiving 5'-DFUR (28 patients). The subjects in the group on 5'-DFUR preoperative administration were classified into three sub-groups according to total dosages of 400 mg (8 patients), under 4,800 mg (8 patients) and 4,800 mg or over (9 patients). In terms of tumor diameter, significant increases (p = 0.027) of S-phase were observed in patients with the diameters of 5 cm or more in the sub-group of under 4,800 mg dosage. Similar tendencies were seen in all drug administration groups (p = 0.05). However, no influence was observed on S-phase in the group without 5'-DFUR administration. In terms of ER stratification, ER(-) patients in the dosing groups showed increasing tendencies in S-phase and significant differences of G1- as well as G2+M phases. No differences in PyNPase activity were noted in each stratified group. Preoperative administration of 5'-DFUR resulted in alterations of cell cycle in breast cancers. In patients with a total drug dosage of under 4,800 mg and with tumor diameters of 5 cm or more, increases of S-phase and decreases of G1-phase were observed. It was suggested that 5'-DFUR acted in DNA synthesizing stage to accumulate S-phase inhibiting cell cycles. Consequently, this drug is considered useful for neoadjuvant chemotherapy.

The relationship between nuclear DNA ploidy pattern and cytological atypia in aspiration cytology of breast cancer was studied in 40 cases of primary breast cancers (T1: 14 cases; T2: 26 cases). They were divided into two groups, according to cytological specimens, high degree (H group) and low degree (L group). Twelve of 40 cases were classified into H group and 28 cases into L group. Fresh frozen materials were obtained from mastectomy specimens and nuclear DNA analysis was carried out by flow cytometry (FCM). Estrogen receptor (ER) and progesterone receptor were measured with the DCC method. Lymph node metastases were histologically seen in 17 cases; the ER positive rate was 55.0%, and the PgR positive rate was 42.5%. FCM revealed 21 cases (52.5%) with diploidy pattern and 19 cases (47.5%) with aneuploidy pattern. S-phase fraction in 17 cases (48.6%) exceeded 20%. L group had more diploid cases and ER positive cases than H group (p < 0.05, p < 0.01). No differences in other factors were seen between the two groups. In diploid cases, L group had more ER positive cases than H group (p < 0.02).

To study the cell characteristics of relatively small pancreatoduodenal cancers (less than 4 cm in the greatest dimension in histopathological examination), we analyzed the relation between the cell kinetics, and histopathological and immunohistochemical findings. Tumors were classified into t1 (0-2 cm in its greatest dimension) and t2 (2-4 cm). We analyzed DNA contents of five tumors of t1 and twelve of t2 by cytofluorometry. Histopathological and immunohistochemical (CA19-9 and CEA) examinations were also carried out. Nuclear DNA content histograms of the tumors in the t1 group showed all diploid pattern, while those in the t2 group exhibited 4 diploid, 7 euploid polyploid and 1 aneuploid patterns. Invasions to the lymphatic (ly), vein (v) and nerve (pn) were found in five cases of diploid tumors (3 of t1 and 2 of t2) and all (8 in t2) of non-diploid tumors. A medullary pattern of tumor growth was only noted in three tumors of t1. This pattern was not found in t2 (4 diploid, 8 non-diploid). These results suggest that the cellular DNA content increased in association with the tumor growth in size. In the histopathological findings, both the invasions (ly, v, pn) and fibrosis of the stroma more often appeared as the tumor size and polyploid and aneuploid cells increased. However, other histopathological findings and the immunohistochemical examination for CA19-9 and CEA had no relation to the tumor growth and the DNA ploidy pattern.

Using flow cytometry, we evaluated the diagnostic usefulness of DNA aneuploidy in epithelial elevated lesions of the stomach. Four biopsy specimens were obtained from each lesion in 14 elevated early cancers, 21 adenomas, 37 hyperplastic polyps and 11 cases of chronic gastritis with intestinal metaplasia. All samples obtained from benign lesions indicated diploidy. DNA aneuploidy was detected in 12 lesions (85.7%) of early cancers. DNA indices ranged from 1.10 to 1.19 in six of 10 intramucosal cancers with DNA aneuploidy. Meanwhile, a total of two submucosally invasive cancers had stemlines with a DNA index of more than 2.00. Four early cancers, including two submucosally invasive ones, had multiple aneuploid stemlines. DNA aneuploidy is a useful marker of malignancy in epithelial elevated lesions of the stomach. A higher DNA index and multiple aneuploid stemlines may be possible markers of submucosal invasion in early cancer.

Neoadjuvant chemotherapy with CDDP and UFT was performed on far advanced gastric carcinoma cases whose curative resection was impossible. These trials were carried out on 10 patients who had Borrmann 4 type carcinoma. PR was found in 6 cases, for an efficacy rate of 60% (6/10). The DNA content in gastric carcinoma was determined before and after treatment by flow cytometry. The relationship between the change of DNA ploidy pattern and effects of chemotherapy was investigated. Aneuploidy was found in 6 of 10 patients (60.0%). DNA ploidy patterns showed no correlation with effectiveness. In aneuploid cases, the group of responders showed seemingly the change of ploidy pattern, from aneuploid to diploid. In non-responders the percentage of aneuploid cells increased. In diploid cases, an accumulation of S-phase fraction was observed in the group of responders. These results suggested that DNA ploidy analysis may provide additional diagnostic criteria for better assessment of chemotherapeutic effects.

An immunohistological staining for p53 conducted in 114 cases of colorectal cancer. The background risk factors such as pathological and DNA ploidy patterns were analized in all cases. p53 was found in 50.0% (57/114). There was a slight increase in p53 positive cases in the advanced cancers, but it was not statistically significant. Five-year survival rates of Dukes B and C were 42.2% for p53-positive cases and 67.2% for p53-negative cases. There was a statistical difference (p < 0.05), but none in the Dukes A group. Study of p53 in combination with DNA ploidy patterns revealed an even greater disparity in survival rates (p < 0.01) between p53-positive aneuploid cases and p53-negative diploid cases. These results indicate that the staining of p53 in conjunction DNA ploidy patterns may be a useful indicator of prognosis in the cases of colorectal cancer.

DNA content was measured by flow cytometry using paraffin-embedded material from 148 primary rectal cancers and 10 distant metastatic lesions. DNA ploidy pattern was classified into the following three groups in terms of the DNA index: 1.00 was defined as the diploid group (DP), 1.00 < DNA index < 1.60 as the low-aneuploid group (LAP) and > or = 1.60 as the high-aneuploid group (HAP). Primary carcinomas were DP in 26.4%, LAP in 36.5% and HAP in 37.1%. The DNA ploidy pattern of the primary tumor correlated well with clinicopathological findings such as depth of invasion, lymphatic invasion, pathological stage, metastasis to distant organs and curability of the tumor. In patients with HAP tumor after curative operation, the recurrence rate (21.6%) in distant organs was significantly higher than those with DP tumor (2.8%) and LAP tumor (6.4%) [p < 0.05]. Two LAP patients and 8 HAP patients with distant metastatic disease had the same DNA ploidy pattern HAP in the metastatic lesions. These data indicate that tumor DNA ploidy patterns classified into three groups in rectal cancer may play an important role in predicting prognosis, including distant metastasis.

Patients with DNA diploid tumor have a significantly longer survival than those with DNA aneuploid in colorectal cancer, but some patients with diploid tumor show poor survival. For assessing malignant potential in diploid colorectal cancer, flow cytometric analysis of S-phase fraction (SPF) was investigated using paraffin-embedded materials from 79 carcinoma patients who had been treated from 1971 to 1989. There was no significant correlation between SPF and clinicopathological factors (such as sex, tumor size, tumor location, macroscopical type, depth of invasion, node metastasis, peritoneal dissemination, liver metastasis and clinical stage). Mean SPF was significantly higher in poorly differentiated type carcinomas than in well differentiated type carcinomas. Patients with higher SPF had a tendency to poorer survival than those with lower SPF. From these results, it was concluded that SPF represents histological differentiation and that SPF may be a prognostic indicator, in diploid colorectal cancer.

Flow cytometric measurement of nuclear DNA content in 159 colorectal adenomas was carried out to investigate the relationship between DNA ploidy and the histological findings. DNA aneuploidy was detected in 18 lesions (12.8%). The incidence of DNA aneuploidy was significantly higher in tubulovillous adenomas than in tubular adenomas (30.4% vs. 8.1%; p < 0.01). DNA aneuploidy was not found in any adenoma with mild dysplasia, but was noted in 19.1% of those with moderate dysplasia and in 33.3% of those with severe dysplasia. The mean size of the lesions was significantly larger in adenomas with aneuploidy than in those without aneuploidy (14.0 mm vs. 7.7 mm; p < 0.01). The DNA index values of 18 adenomas with aneuploidy were divided into two groups: one ranged from 1.07 to 1.23 and the other from 1.66 to 1.85. DNA index values correlated with the size of the lesions (p < 0.05), but not with the histologic type and degree of dysplasia.

Amplification of hst-1 gene is associated with poor prognosis in patients with esophageal carcinoma. Since there is a high frequency of DNA stem-line heterogeneity, we studied intratumoral heterogeneity of hst-1 amplification, and evaluated whether or not the intensity of hst-1 amplification is associated with the malignant potential of esophageal carcinoma. A total of 73 patients with esophageal carcinoma who had undergone esophagectomy were studied. Intratumoral heterogeneity of hst-1 amplification was examined in two to four sections of tumour specimens which showed mainly DNA heterogeneity in 27 of these patients. The judgement of hst-1 amplification in the same tumor was identical in all of the 27 patients, although its intensity was not identical in some cases. And in 73 patients, postoperative recurrence in organs showed a high incidence (78%) in the hst-1 amplification group with a high intensity (over 7-fold). Therefore, hst-1 amplification exceeding 7-fold can serve as an indicator to predict the high grade of malignant potential of esophageal carcinoma.

Numerical aberrations of chromosomes can be detected by fluorescence in situ hybridization (FISH), using chromosome-specific probes. It is possible to observe this in solid tumors from which it is very difficult to obtain metaphase nuclei. The present study employed surgical specimens from 15 cases of colorectal carcinoma, all of which showed DNA diploidy. In the same samples, we analyzed chromosomal numerical aberration by FISH according to the method of Pinkel et al. Biotinylated DNA probes specific to chromosome #7, #11 and #17, were used. The hybridization spots were observed by fluorescent microscopy. As a result, the numerical aberrations of chromosomes detected by FISH were found in DNA diploid cases by FCM. They were trisomy and monosomy. These results indicate that FISH is useful to detect the chromosomal aberrations of DNA diploid cases.

Image analysis of the nuclear morphology was performed on propidium iodide-stained isolated cells from 9 adenomas and 9 cancers of the colorectum. We analyzed DNA content, seven nuclear geometric features and four shape factors of tumor cells using an image cytometry system that has been developed in our laboratory. Nuclear breadth and the degree of contour irregularity of cancer cells were found to be significantly different from those of adenoma cells. The coefficients of variation (CV) of some nuclear features and the degree of contour irregularity were increased in cancer cells, while CV of the degree of circularity was increased in adenoma cells. Moreover, nuclear in G2 phase were found to be larger and more round than those in G1 phase. These results suggest that cell-cycle-related analysis of nuclear features would be a valid means to assess the nuclear morphology of colorectal tumors.

From retrospective analysis of 174 formalin fixed, paraffin-embedded, HCC tissue samples, the possible correlation was investigated between the CV determined by FCM and the duration of formalin fixation. An adverse effect of prolonged formalin fixation on CV and fluorescence intensity in liver tissue was investigated prospectively in samples from 4 patients.

Employing surgically resected specimens from patients without preoperative radiation therapy, we measured the nuclear DNA contents in order to evaluate the malignant potential of esophageal carcinoma. The association of the DNA index (DI) > or = 1.7 and the intratumoral DNA heterogeneity of biopsy specimens from patients with or without preoperative radiotherapy was also analyzed in relation to problems related to preoperative evaluation of nuclear DNA contents using biopsy specimens and the influence of irradiation on nuclear DNA contents. Esophageal carcinomas in 128 cases were studied. Nuclear DNA content was measured for biopsy specimens as well as surgically resected tumors by flow cytometry according to the method of Hedley and colleagues. HET was found in 42% of surgically resected specimens, but in only 18% of biopsy specimens. In 90% of cases, dominant DI of the resected tumor was also found in the biopsy specimens. Cases of DI > or = 1.7 showed a poorer prognosis than those with DI < 1.7 in cases with preoperative radiotherapy (p < 0.05) as well as in those without it. These results indicated that DI can be an indicator for highly malignant potential of esophageal carcinoma when measuring nuclear DNA contents using not only biopsy specimens but also specimens from irradiated cases.

The present study was undertaken to examine the influences of excessive contamination by lymphocytes on the flow cytometric analysis of DNA ploidy of head and neck cancers. Although lymphocytes which contaminate test cells can serve as an internal control (2C), an excess of these lymphocytes can mask the aneuploid peak of cancers. This tendency was particularly prevalent in well-differentiated squamous cell carcinoma. In some cases, most of the cells isolated from tumor-positive HE-stained tissue samples were lymphocytes. This is probably because lymphocytes are likely to survive the cell-isolating procedure, whereas keratocytes are likely to be destroyed during it. If samples excessively contaminated by lymphocytes are subjected to flow cytometry, the aneuploid peak of cancer cells is low relative to the 2C peak, possibly causing an erroneous judgment of diploid. Cytofluorometry, in which a smear of test cells is observed directly to quantify the DNA of cancer cells alone, was useful in detecting lymphocyte-masked aneuploid cells.

A total of 9 surgically resected non-cancerous benign lesions (5 regenerative epithelium, 4 hyperplastic or metaplastic epithelium), 4 normal epithelium, 30 carcinomas and 12 adenomas containing 7 moderate dysplastic epithelium and 5 severe dysplastic epithelium in the large intestine, were used. Then 40 microns thick paraffin sections were made from each specimen, dewaxed, and the target portion was excised under microscopic visualization. Free cells isolated by 0.5% pepsin solution were smeared and followed by DNA staining by DAPI, and fluorocytometric DNA measurement was performed using a fluorocytometer (BH 2.QRFL, Olympus). All specimens of non-cancerous benign epithelium and normal epithelium showed diploidy. The average mean DNA value was 2.03 +/- 0.04 c (1.95-2.10), the percentage of cells above 4.1 c (4.1 c > or = %) averaged 0.26 +/- 0.42% (0-1.25) and the coefficient of variation (CV) averaged 14.30 +/- 3.18% (10.19-19.58). Seven of 30 carcinomas showed diploidy. Among these diploid carcinomas, the mean DNA value averaged 2.30 +/- 1.98 c (2.01-2.71) (p < 0.05), 4.1 c > or = % averaged 4.17 +/- 2.91% (2.0-10.5) (p < 0.05) and CV averaged 32.2 +/- 5.8% (24.43-38.90) (p < 0.01). Based on these results, aneuploid cases were judged to be malignant. In the case of diploidy, if the percentage of cells above 4.1 c was more than 2.0% and CV was more than 24.43%; when the former was less than 1.25% and the latter was less than 19.58%; and when one or both of the two was out of the range described in the above two categories, they were judged to be malignant, benign or borderline, respectively. From these criteria, 7 adenomas showing moderate dysplasia were analyzed as benign in 2, borderline in 4 and malignant in 1, while 5 adenomas evidencing severe dysplasia were analyzed as malignant in 4 and borderline in 1. In conclusion, fluorocytometric evaluation of the malignant or benign tumor of borderline lesions in the large intestine was found to be possible.

DNA ploidy of 45 smooth muscle tumors of the G.I. tract was determined by flow cytometry and correlated with clinical features and prognosis. The sites of the tumors were: esophagus (1), stomach (24), small intestine (12), large intestine (6), liver (1) and pancreas (1). The histologic type was leiomyoma in 14, leiomyosarcoma in 29, and leiomyoblastoma in 2. DNA aneuploidy was more frequent in leiomyosarcoma (17/29) than leiomyoma (5/14), but the difference was not statistically significant. One leiomyoblastoma was diploid and the other was aneuploid. No patients with leiomyoma died. In patients with leiomyosarcomas, 5-year survival was significantly poorer in those with aneuploid tumors (38%) than in those with diploid tumors (83%). There was no correlation between DNA ploidy and clinico-pathological features of tumors. The present study disclosed that DNA ploidy is a prognostic variable, independent of other variables.

Histopathologic studies and fluorocytometric DNA analysis were carried out in 39 patients with intestinal metaplasia (IM) of the mucosa associated with gastric carcinoma, in order to determine possible relationships. According to the histological degree of IM, 39 cases were divided into 16 (41%) with slight-to-moderate, and 23 (59%) with severe IM. The 39 tumors were histologically divided into 20 (51%) differentiated tumors and 19 (49%) undifferentiated tumors. The DNA patterns of tumors revealed 17 (43%) to be euploidy and 22 (57%) to be aneuploidy. The DNA content of the metaplastic epithelial cells was measured, and the percentage of the cells in S and G2 + M phases of the cell cycle was taken as the index of proliferative activity. DNA analysis of the non-metaplastic epithelium was performed in 10 as a control group. While the histological differentiation of the tumors and the degree of IM were not related to the proliferative activity of IM, the IM around the aneuploid tumors did have higher proliferative activity (14.01 +/- 6.16%) than that around euploid tumors (11.08 +/- 7.13%) or of normal mucosa (9.95 +/- 9.92%). This approach will help to determine cell kinetics of IM around the gastric tumor. Our study indicated a relationship between DNA ploidy and the proliferative activity of IM around gastric cancer.

Biological characteristics of cancer cells are thought to be immortalization, transformation (abnormal cell growth) and capacity for tumor invasion and metastasis. Recent studies have revealed that many cellular genes (oncogenes and anti-oncogenes) were activated or suppressed at the level of DNA in cancer cells, and in normal condition the products of these genes have important roles in signal transduction for cell growth as a positive or negative way. Point mutation and unphysiological DNA recombination are considered to be major molecular mechanisms in the alteration of these cellular genes. Further, several genes were found to be involved in invasion and metastasis. However, still a significant portion of cancers remains to be elucidated in terms of unknown cellular (or viral) genes involved in multi-step carcinogenesis. Therefore, we need to continue the basic research on cancer and to initiate the studies on new chemicals, hormones or others by which we could control the activities of known oncogene and anti-oncogene products.