A fast and reliable system for microcomputerized DNA cytofluorometry in tumour pathology

Cytometric and histopathologic features of tumors detected in a randomized mammography screening program: correlation and relative prognostic influence

Cytometric determination of S-phase fraction and ploidy type was performed on 430 tumors detected within a randomized trial of mammographic screening. The results were compared to several histopathologic features. A high S-phase fraction was estimated in tumors with a high grade of malignancy and other histopathologic findings related to rapid tumor progression, including lack of tubule formation, a high mitotic index, marked nuclear pleomorphism, multifocal cancer growth, tumor emboli in lymphatic and blood vessels, tumor necrosis, and inflammatory reaction. DNA aneuploidy was correlated with a high malignancy grade, frequent mitoses, a high degree of nuclear pleomorphism, vascular invasion, necrosis, and the presence of noninvasive ductal carcinoma. Both cytometric variables were inversely related to the degree of elastosis. Positive nodes, large tumor size, DNA aneuploidy, a high S-phase fraction, high grade of malignancy, lack of tubule formation, as well as high mitotic index and pleomorphism, presence of multifocal cancer, and vascular invasion, predicted a significantly shorter distant recurrence-free interval after a median follow-up time of 46.6 months. Elastosis and the presence of estrogen and progesterone receptors indicated favorable prognosis. In the multivariate analysis, only lymph node status, tumor size, S-phase fraction, and multifocal growth pattern had independent prognostic value.

Influence of S-phase fraction on metastatic pattern and post-recurrence survival in a randomized mammography screening trial

Using static cytofluorometry, S-phase was determined on the primary tumors of 421 patients with breast carcinomas in stages I-III diagnosed 1981-85 during the second and third screening rounds of a randomized trial evaluating the effect of mammographic screening. Through December 1988, 82 patients had developed local and/or distant recurrence, 51 of whom had died of cancer during the same period. The distribution among sites of recurrence differed between patients with tumors detected by mammography screening and cancers diagnosed due to clinical symptoms. The mean S-phase fraction was highest in patients with liver or brain metastases and lowest in patients with metastases in subcutaneous and cutaneous tissue and lymph nodes only. In univariate analysis, survival after first recurrence was significantly associated with the site of primary recurrence, the disease-free interval, and node status and tumor size at diagnosis, as well as the S-phase level. The median survival period was 31.3. months for patients with a S-phase fraction below 6%, and 10.7 months in cases with S-phase exceeding 10%. Neither ploidy nor the estrogen receptor content had significant influence on post-recurrence survival. In Cox's multiple regression analysis, only metastatic site, disease-free interval, and S-phase fraction showed significantly independent prognostic value.

Cytometric characterization and clinical course of breast cancer diagnosed in a population-based screening program

A randomized controlled trial evaluating mammographic screening was started in two Swedish counties in 1977. In one of these, Ostergötland county, the authors performed static cytofluorometry on 161 cancers detected at the second and third screening rounds, 50 interval cancers, and 219 cancers appearing in the nonscreened control group during the same time period. The median follow-up time was 42 months. No difference in mean S-phase was found between screening and control group cancers, but interval cancers, appearing between two screenings, had increased mean S-phase levels (P = 0.01) compared to both of the other groups. A high S-phase fraction was associated with distant recurrence in both node-negative and node-positive tumors. Aneuploid tumors were more often found in the control group (67%) and among interval cancers (72%) than among screening detected cancers (55%, P = 0.02). In Cox's multivariate analysis, including all patients, the lymph node status, tumor size, estrogen receptor content, and S-phase all contributed independent prognostic information about the clinical course. DNA ploidy predicted the outcome in simple but not in multivariate Cox's analysis. When analyzing screening-detected cancers separately, only the S-phase significantly predicted distant recurrence in multivariate analysis. In tumors with local recurrence, a high S-phase implicated an increased, although not statistically significant, risk for distant recurrence. Survival with metastatic disease was significantly influenced by the S-phase level (P = 0.002). The authors conclude that S-phase fraction provides valuable kinetic information related to the clinical outcome for all stages of the disease and serves as a prognostic factor in screened populations, which have tumors predominantly in early stages.

DNA ploidy in oral cavity carcinomas, with special reference to prognosis

Single-cell DNA cytofluorometry was performed on paraffin-embedded tissue of 140 patients with squamous cell carcinomas of the oral cavity. Half of the tumors (71 of 140) were DNA nondiploid. Well-differentiated carcinomas were more often DNA diploid than moderately well-differentiated ones (P less than 0.001; chi-square). The aneuploid tumors responded better to preoperative radiotherapy than did the DNA diploid (P less than 0.001) and polyploid tumors (P less than 0.05; chi-square). Using the multivariate Cox's regression analysis multiploid type tumor, age of the patients and presence of lymph node metastases were the only significant factors influencing survival. DNA diploid tumors in stages I and II had a better prognosis than DNA nondiploid (P less than 0.01; Kaplan-Meier). The reverse was true for stages III and IV, where DNA nondiploid tumors had a better prognosis (P less than 0.05; Kaplan-Meier). Tumor stages (P less than 0.001; Kaplan-Meier) and especially lymph node metastases (P less than 0.0005; Kaplan-Meier) were major prognostic factors. Tumor DNA ploidy may be a complement to clinical and morphologic parameters as a prognostic predictor in squamous cell carcinoma of the oral cavity.

Nuclear DNA distribution in neuroendocrine gastroenteropancreatic tumors before and during treatment

The nuclear DNA contents of tumor cells in 73 patients with endocrine gastrointestinal tumors, 19 patients with endocrine pancreatic tumors (EPT) and 54 patients with malignant carcinoid tumors were determined before and after treatment. The DNA profiles were divided into diploid and aneuploid. In untreated patients, 9 out of 10 (90%) primary EPT and all 9 primary malignant carcinoid tumors (100%) were diploid. Tumor cell imprints from liver metastases of patients with untreated EPT showed aneuploidy in 5 of 11 cases, but only in 7 out of 46 DNA records from patients with untreated carcinoid liver metastases. DNA alteration from diploid to aneuploid profiles occurred in 2 patients with endocrine pancreatic tumors who had received chemotherapy. A change from diploid to aneuploid records was also seen in 7/23 (30%) carcinoid tumors after treatment. The DNA patterns before and after treatment did not show any correlation with survival or treatment response.

A rapid system for static cytofluorometry enabling the simultaneous determination of nuclear size and DNA content

A measuring technique enabling the simultaneous determination of nuclear size and DNA content by single cell cytofluorometry is described. The system is based on a Leitz MPV cytophotometer, a Zeiss scanning stage and a microcomputer system. Cells to be measured are not positioned in the ordinary way, but are passed through the excitation light beam. In this way rapid measurements are achieved. Further, an estimate of the object diameter is calculated from the pulse shape obtained as the fluorescence is recorded continuously by using a circular measuring aperture, congruent with the excitation light field. Repeated measurements of nuclei stained with Hoechst 33258 showed a high reproducibility for DNA content with a coefficient of variation (CV) below 1% and an acceptable precision for nuclear area with a mean CV of 7%. Nuclear area estimated by the method was well correlated to area determined from photographs (r = 0.986). Disintegrated samples from eight patients with breast cancer were analyzed and compared to Feulgen-stained samples analyzed by scanning cytophotometry. Scatter diagrams illustrating DNA content and nuclear size showed essentially the same distribution with the two methods. We conclude that the system may be a rapid alternative when fluorescence together with object size are quantified in slide preparations.

Comparison of static and flow cytofluorometry for estimation of DNA index and S-phase fraction in fresh and paraffin-embedded breast carcinoma tissue

DNA index, S-phase and G2/M fractions were measured in breast carcinomas before and after paraffin-embedding. Unfixed tumors were analysed with flow cytometry, while cells prepared from fixed and paraffin-embedded tumors were analysed with both static cytometry and flow cytometry. Cells for static cytometry were prepared with pepsin and cells for flow cytometry were prepared with trypsin. In 44 of 51 paraffin-embedded tumors, analysed with flow cytometry, both DNA index and S-phase fraction could be estimated. There was a strong correlation between the three methods of obtaining DNA indices and S-phase fractions, but this was not true for G2/M fractions. Histograms from flow cytometry of archival material were corrected for background. We conclude that cells prepared from formalin-fixed and paraffin-embedded breast cancer tissue can yield reliable estimation of DNA index and S-phase fraction.

Prognostic factors in locally advanced breast cancer (T3, T4) with special reference to tumor cell DNA content

The prognostic value of DNA analysis was studied retrospectively in 91 patients with locally advanced breast cancer (T3, T4) and a follow-up time of 3-7 years. Tumor cell DNA analysis was performed by static cytometry on aspiration biopsy specimens in 42 cases and on tissue sections in 49 cases. The tumors were classified as euploid or aneuploid. Sixty-four percent of the tumors were aneuploid. DNA pattern correlated significantly to histologic grade and axillary perinodal growth and also to survival. DNA pattern, histologic grade and axillary node metastases correlated significantly to disease-free survival (DFS). In this series of patients with locally advanced breast cancer DNA determinations gave important prognostic information.

DNA content, malignancy grading and prognosis in T1 and T2 oral cavity carcinomas

Microscopic malignancy grading using the 8-factor system proposed by Jakobsson et al. (1973), the 4-factor system set up by Glanz and Eichhorn (1985), and DNA cytofluorometry were applied to thirteen T1 and thirty-seven T2 squamous cell carcinomas of the oral cavity, 9 with and 41 without metastases. There was a significant correlation between the presence of lymph node metastases (N1) and the malignancy scores (P less than 0.05) and tumour DNA ploidy (P less than 0.01, chi-square). The total number of patients with initial and late lymph node metastases correlated significantly with polyploid nuclei (P less than 0.05) and with malignancy scores (P less than 0.001), which also correlated with regional recurrences (P less than 0.01, chi-square). No remaining tumour after preoperative radiotherapy indicated less risk for local recurrence than if tumour persisted (P less than 0.01, chi-square). The cumulative survival (Kaplan-Meier) was worse for patients with nodal involvement (N1) than for those without (N0) (P less than 0.01), and for patients with poorly differentiated tumours compared with moderately well differentiated (P less than 0.05) and to well differentiated (P less than 0.001). The prognosis was worse for patients with high malignancy scores than those with low (P less than 0.001). DNA diploid tumours had a better prognosis than DNA non-diploid, but the difference was not significant.

DNA cytophotometry of acinic cell carcinoma and its relation to prognosis

Single-cell DNA cytophotometry has been assessed on histological sections or aspiration biopsy smears from 13 parotid acinic cell carcinomas and has been related to prognosis. The minimal follow-up was 6 years. Six ACC were well differentiated and seven moderately well differentiated. The modal DNA values of all tumours examined were in the diploid area. The number of nuclei above 2.5C DNA, expressing proliferative ability or aneuploidy, was low. A difference, too small to be of value for prognostic purposes, was found between the non-recurrent tumours (6%) and the metastasizing tumours (16%). The DNA levels measured in the present tumour material correspond to the concept of ACC as being a well differentiated neoplasm. The nuclear DNA pattern of these tumours would seem to be of minor prognostic significance with regard to survival time and/or potential for metastasis.

Nuclear DNA recordings in gastric carcinoids. A cytofluorometric study on single tumour cells

Cytofluorometric nuclear DNA analyses were performed on single tumour cells from eight argyrophilic gastric carcinoids. Three tumours displayed grossly diploid DNA values, with a fraction of S-phase cells of 1-8%. Four other tumours were also mainly diploid, but showed an increased number of tetraploid cells, and one tumour was aneuploid with a stem-cell line of triploid cells. These and previous results indicate that gastric carcinoids are relatively heterogenous with respect to their morphology, nuclear DNA profiles and biological behaviour, in contrast to most intestinal "classical" carcinoids. Furthermore, the results are in accordance with the suggestion that gastric carcinoids may represent a highly differentiated variety within a spectrum of neuroendocrine neoplasia analogous to the corresponding variety observed in the bronchus (carcinoid-small cell, undifferentiated carcinoma).

Malignancy grading and cytophotometric evaluation of T1 lip carcinomas

Histological malignancy grading using the score system set out by Jakobsson et al. in combination with static DNA cytofluorometry was applied to 20 T1 squamous cell carcinomas of the lower lip, 11 with and 9 without lymph node metastases. The mean malignancy score was 18.0 for tumours with metastases and 13.0 for those without (P less than 0.05, chi square test). Nuclear polymorphism, mode of invasion, vascular invasion, and cellular response were the single factors with the strongest influence. All but 3 carcinomas were DNA diploid, but DNA ploidy and proliferative activity (S-phase) yielded no prognostic information. However, hypertetraploid cells were found in all DNA diploid carcinomas with metastases, but in only half of those without (P less than 0.05).

DNA pattern in oral cavity carcinomas in relation to clinical stage and histological grading

Cytofluorometric DNA analysis was performed in 88 squamous cell carcinomas of the oral cavity. 48% (42/88) of the tumours were DNA non-diploid. The frequency of DNA non-diploid tumours seemed to correlate with the increasing size of the tumour, and the decrease of histological grading, and certainly with the presence of lymph node metastases (p less than 0.001). DNA non-diploid tumours often had shorter duration of symptoms. The S-phase level seemed to increase the less differentiated the tumour but did not correlate either to tumour size or to duration of symptoms. Polyploid nuclei were more common in poorly differentiated tumours (9/10) as compared to well differentiated ones (17/33) (p less than 0.05) and seemed to be more common in tumours with metastases (20/26) than in those without (40/62).

Preoperative irradiation in oral cavity carcinoma. A study with special reference to DNA pattern, histological response and prognosis

The relation between DNA pattern and response to preoperative radiotherapy was studied in 51 patients with oral squamous cell carcinomas operated 4 weeks after irradiation. Small tumors (T1 and T2) showed more pronounced response to radiotherapy than larger ones (T3 and T4), as did DNA aneuploid tumors. Eight of 11 DNA aneuploid tumors showed no remaining tumor in the operation specimen, compared to 6 of 21 DNA polyploid and 2 of 19 DNA diploid tumors. None of 16 patients without demonstrable remaining cancer in the operation specimen had local recurrence. The presence of lymph node metastases was the most important prognostic factor.

Prognostic value of deoxyribonucleic acid content in metastatic renal cell carcinoma

The deoxyribonucleic acid content of tumor specimens from 23 patients with metastatic renal cell carcinoma was analyzed prospectively by flow cytometry and static cytophotometry. Of the primary tumors 10 (43 per cent) were homogeneously diploid or near diploid in 8 samples studied from each tumor and 13 (57 per cent) had an aneuploid deoxyribonucleic acid content in 1 to 8 samples. At the end of followup 9 of 10 patients with diploid or near diploid primary tumors were alive, compared to only 1 of 13 with aneuploid primary tumors. Patients with homogeneously diploid or near diploid tumors survived significantly longer compared to those with aneuploid tumor deoxyribonucleic acid content (p less than 0.001). After excision of solitary diploid or near diploid metastases 4 patients had no evidence of disease. In 3 of these patients the primary tumors were diploid or near diploid, whereas 1 had 1 aneuploid and 7 diploid or near diploid samples in the primary tumor. In 10 other patients 28 metastases revealed concordance in deoxyribonucleic acid content with the primary tumors. Our results indicate that deoxyribonucleic acid content might be a useful prognostic discriminator with implications for the clinical management of patients with metastatic renal cell carcinoma.

DNA content and prognosis in renal cell carcinoma. A comparison between primary tumors and metastases

Deoxyribonucleic acid (DNA) content was retrospectively determined by single-cell cytophotometry in primary tumors and corresponding metastases from 32 patients with renal cell carcinoma. In 15 of the primary tumors a diploid/near diploid and in 17 an aneuploid DNA content was found. A diploid/near diploid DNA pattern was revealed in 10 metastases and 22 were aneuploid. By comparing the DNA content in the primary tumors with their metastases, 13 of 32 showed a clear divergency, which might illustrate tumor cell heterogeneity of renal cell carcinoma. The DNA pattern showed a close correlation to morphologic grading. A correlation between DNA content in the metastases and survival time was found. Patients, with diploid/near diploid metastases survived significantly longer than those with aneuploid DNA contents (mean, 31.1 and 11.5 months, respectively; P = 0.004). In contrast to this, no correlation was found between DNA content in the primary tumors and survival time.

DNA-measurements on formalin-fixed, paraffin-embedded squamous cell carcinomas from different ENT-regions

DNA measurements were performed on formalin-fixed, paraffin-embedded squamous cell carcinomas, using the method of Hedley et al. Ultrasonication was sometimes necessary after enzymatic disintegration to dispose of residual cytoplasm. The isolated nuclei were stained with Hoechst 33258. DNA measurements were performed using a rapid system for static fluorometry (FLUORA-programme). The method proved to be highly accurate. The mean CV of the stem cell peaks from different tumours was 5.3. There was no difference between material less or more than 5 years old (mean CVs 5.2 and 5.5 respectively). RNase treatment had no effect and was therefore not considered necessary. The proliferative activity was calculated by computer assuming a rectilinear distribution of S-phase cells. Because routinely paraffin-embedded material can be used, DNA analysis can be performed on tumour material where the clinical outcome is known, in order to evaluate certain DNA variables.

DNA measurement--an objective predictor of response to irradiation? A review of 24 squamous cell carcinomas of the oral cavity

DNA measurements on biopsy material from 24 squamous cell carcinomas of the oral cavity given preoperative radiotherapy indicate that DNA aneuploid tumours respond better to radiotherapy than do diploid and polyploid tumours. The mean S-phase value was higher (16.1%) for 8 tumours that were eradicated by preoperative radiotherapy than for 13 that did not respond (8.1%). These factors correlated better with the response than did histological and clinical (T) classifications. DNA-ploidy and S-phase estimation can complement the histological diagnosis, and may prove valuable when planning treatment.

A comparison of static cytofluorometry and flow cytometry for the estimation of ploidy and DNA replication in human breast cancer

332 primary invasive breast carcinomas were analysed by static cytofluorometry and flow cytometry. The ploidy distributions were similar, and 54% of the tumors were judged DNA aneuploid by both methods. The coefficient of variation of the G0-G1 peaks ranged from 2.0 to 8% with both techniques, but the mean was somewhat lower with flow cytometry--4.1%, compared to 4.9% for the static measurements. The proportion of S-phase cells was possible to estimate from 80% of the flow histograms and 70% of the static histograms. S-phase was not estimated from the static histograms if less than 150 tumor cells were measured. With 160 tumors S-phase was measured by both methods. The range was 0 to 27% with the static measurements and 0.7 to 25% with flow cytometry. Corresponding mean values were 7.6% and 8.2%, which are similar to thymidine labeling index results with breast cancers reported in some studies. A close correlation was obtained (r = 0.927) comparing S-phase fractions estimated from aneuploid tumors with flow cytometry and static cytofluorometry if more than 200 cells were measured with the latter. The proportion of S-phase cells was significantly lower for the diploid tumors. We conclude that both techniques can be useful for the estimation of DNA ploidy and replication in human breast cancer.

DNA content in renal cell carcinoma. A comparison between flow and static cytometric methods

Different flow and static cytometric techniques for analysis of DNA content were compared using imprints, fresh and paraffin-embedded tissues from renal cell carcinomas. All methods discriminated between diploid-/near diploid and aneuploid samples and gave comparable results concerning modal ploidy levels (r = 0.744-0.970). For retrospective investigations of renal cell carcinoma, flow cytometry using paraffin-embedded tissues seemed to be the method of choice in determination of DNA content.

Use of formalin fixed, paraffin embedded tumours for estimation of cellular DNA content and S-phase fraction by static cytofluorometry

Sections with a thickness of 50 micron from 34 formalin fixed, and paraffin embedded tumours were deparaffinized and hydrated and the cells were disaggregated with pepsin. The cells were stained with Hoechst 33258 and a rapid method for static fluorometry was used. DNA ploidy and the fraction of cells in S-phase were estimated and compared with results obtained with freshly prepared cells. The coefficients of variation for the tumour stemlines for both the fresh and the embedded material were approximately 6 per cent. There was a good correlation between DNA ploidy determined from fresh and embedded material, respectively; there was also a close correlation between the fractions of S-phase cells estimated from fresh and embedded tumours. It was also possible to cut out small histologically defined lesions, such as carcinoma in situ of the breast from thick paraffin sections and obtain DNA histograms with high resolution.

DNA content in renal cell carcinoma with reference to tumor heterogeneity

DNA content was successfully determined by flow cytometry in 196 tissue samples from 25 renal cell carcinomas. Twelve tumors (48%) were homogeneously diploid/near-diploid, whereas 11 tumors were aneuploid and 2 tumors were polyploid. Cell clones with different DNA content were found in 11 tumors demonstrating a considerable heterogeneity in the non-diploid tumors; 9 of these 11 heterogeneous tumors contained both aneuploid and diploid cell clones. Tumor samples morphologically classified as grade 1 and 2 were 98% diploid and grade 4 samples were 78% aneuploid. No correlation between DNA distribution and morphologic grade was found for grade 3 tumor samples. Tumor proliferation rate, as determined by the fraction of cells in S-phase, was significantly higher in aneuploid samples compared to normal kidney tissue samples and diploid tumor samples.

Receptors for retinoic acid and retinol in human mammary carcinomas

The cellular content of receptors for retinol (CRBP) and retinoic acid (CRABP) was measured in 148 human mammary carcinomas. High levels of CRABP were found in lobular carcinomas while those of the papillary subgroup had low levels of the receptor. Intermediate values for CRABP were observed for ductal, colloid and medullar carcinomas. The cellular levels of CRBP were high in ductal carcinomas and low in papillar carcinomas. A positive correlation was observed between the receptor for estradiol and CRABP. However, no significant correlation was found between the tumor cell DNA pattern and the content of CRABP and CRBP respectively. Recurrence of disease could be predicted by the nodal status and the level of estradiol receptor while the DNA pattern and the content of receptors for retinoic acid and retinol failed.

3H-thymidine autoradiography and cytophotometric analysis of needle aspirates from human tumours during radiation therapy, endocrine therapy and chemotherapy

Cell kinetic variables in human tumours were investigated before and during therapy. Tumour cell material was collected by sequential thin needle aspiration biopsy. Small variations in DNA replication could be detected by 3H-thymidine autoradiography. In tumours with a high rate of replication, therapy induced changes could also be detected with flow cytometry and static cytophotometry. Endocrine treatment with glucocorticoids, radiation therapy, and chemotherapy were associated with reduced rates of DNA replication in malignant lymphomas. The conformity between labelling index and S-phase estimates from DNA histograms was poorer in biopsy specimens collected during therapy than in pretreatment specimens.

An evaluation of Feulgen-acriflavine-SO2 and Hoechst 33258 for DNA cytofluorometry in tumour pathology

The performance of Feulgen-acriflavine-SO2 and Hoechst 33258 for cytofluorometric ploidy determination is evaluated and compared. The fast initial fading of acriflavine-SO2 stained nuclei, with substantial reduction in fluorescence intensity within about 10 ms, is shown to be related to changes in DNA-ratios between cell types when measurements are performed using different excitation light intensities. Although low intensity excitation-, without such fading-, was used, the acriflavine-SO2 staining procedure showed large specimen to specimen variability regarding both mean fluorescence intensities and coefficients of variation (CV). The cells in such specimens could be shown to contain the same amount of chromophore using scanning absorbance measurements and the differing fluorescence values therefore probably represent variability in fluorescence yield. Specimens stained with Hoechst 33258 after RNase treatment showed very small deviations in mean fluorescence value between slides, and also much lower CV's than in acriflavine-SO2 stained specimens. The results indicate that with proper internal standards this procedure would allow the detection of deviations in DNA content in the order of 2%.