Expression of the ALK protein by anaplastic large-cell lymphomas correlates with high proliferative activity.

A variable fraction of anaplastic large-cell lymphomas (ALCLs) exhibits a t(2;5)(p23;q35) translocation that results in expression of the chimeric hyperphosphorylated protein NPM-ALK (p80). Tumor cells expressing NPM-ALK exhibit markedly enhanced proliferative activity, but comparative cellular kinetic studies on ALK(+) (ALK lymphomas) and ALK(-) lymphomas are lacking. The present study showed that ALK(+) lymphomas, detected with the monoclonal antibody ALKc (n = 17), had significantly higher average values for the proliferation-associated parameters mitotic index, ana/telophase index, growth index (x x mitotic index - apoptotic index, assuming x = 3), percentages of Ki-67(+) cells and fraction of cells expressing cyclin A or B or the cell cycle-regulatory protein p34(cdc2) than did ALK(-) ALCLs (n = 15). Whether this intense proliferative activity contributes to the good response to chemotherapy and favorable outcome of ALK(+) ALCLs remains to be assessed in a larger series of patients. Our findings support the notion that ALK(+) and ALK(-) ALCLs are 2 distinct disease entities.

Immunogold visualization of actin near the coated pits at the apical surface of human mammary epithelial cells.

This ultrastructural study of both the normal human breast tissue and differentiated mammary carcinoma (NOS) epithelial cells has revealed pictures demonstrating luminal receptor-mediated endocytosis. By application of immunogold anti-actin labeling, actin surrounding the fusion ring of coated pits was visualized. However, the coated membrane was not actin labeled. We suggest that association of the actin with coated pits may evidence for its participation in pinching off of the coated vesicles.

Expression of p34(cdc2) and cyclins A and B compared to other proliferative features of non-Hodgkin's lymphomas: a multivariate cluster analysis.

In view of recent knowledge on proteins regulating the cell cycle, we re-evaluated proliferative features of 98 diffusely growing non-Hodgkin's lymphomas. The combined use of 5 proliferation-associated variables (mitotic indices and percentages of Ki-67(+), p34(cdc2+), cyclin A(+) and cyclin B(+) cells) and their entry into a multivariate cluster analysis separated, without overlaps, the entire cohort into 3 groups (clusters) with (1) low, (2) intermediate and (3) high proliferative activity. Conversely, bivariate plots exposed considerable cluster overlaps. Multivariate stepwise discriminant analysis of all cases revealed a decreasing order of discriminant power for % Ki-67(+) cells > % p34(cdc2+) cells > mitotic index > % cyclin A(+) cells > % cyclin B(+) cells. The combined use of 2 variables only, mitotic index and % p34(cdc2+) cells, allowed a clear-cut separation of clusters 2 and 3. In bivariate plots, correlations were best between % Ki-67(+) cells and % cyclin A(+) cells and between mitotic indices and % cyclin B(+) cells. Except for chronic lymphocytic leukemias, immunocytomas and marginal zone lymphomas (all in cluster 1), individual lymphoma entities were distributed among at least 2 clusters. There was, however, a marked preponderance of mantle cell lymphomas and diffuse follicular center lymphomas in cluster 1 and of diffuse large B-cell lymphomas and peripheral T-cell lymphomas in cluster 2. Anaplastic large-cell lymphomas predominated in cluster 3 and responded best to therapy.

Oncocytic (Hürthle cell) tumors of the thyroid: distinct growth patterns comparedwith clinicopathological features.

Neoplastic growth results from cell production that exceeds cell loss. We registered mitotic and apoptotic indices (MI and AI) in 97 immunohistochemically verified oncocytic (Hürthle cell) tumors of the thyroid (OT; 50 adenomas [OA], 20 atypical adenomas [aOA], and 27 carcinomas [OC]) and compared these kinetic data with histological diagnoses and other parameters. MI, although very low in all, was significantly higher in carcinomas than in adenomas. Conversely, AI did not differ as much among the 3 groups. This indicates that the magnitude of cell deletion did not play a prominent role in determining the disparate growth of the 3 types of oncocytic tumors. Cluster analysis with MI and AI per case as variables revealed the existence of 3 groups of neoplasms with highly distinct growth characteristics: (1) near-steady state (n = 78, all diagnostic categories represented); (2) progressive (n = 9, mostly carcinomas); and (3) regressive (n = 10, mostly adenomas). MI distinguished between histologically benign and malignant with the greatest discriminant power of the variables tested. Proliferative indices should thus be included in the differential diagnostic evaluation of oncocytic thyroid tumors. Our study also suggests that invasiveness and growth are 2 diverging properties of carcinomas.

Cellular kinetic differences between Hodgkin's and anaplastic large cell lymphomas: relation to the expression of p34cdc2 and cyclin B-1.

Our study was designed to compare cellular kinetic parameters of classical Hodgkin's disease (HD) with those of anaplastic large cell lymphomas (ALCL-C, common type; and ALCL-HL, Hodgkin's like), with a particular focus on the G2/M transition. These disorders share some phenotypic properties, e.g., CD30 positivity of putative neoplastic cells. The percentages of cells expressing p34cdc2 (p34) and cyclin B-1 (cyclin-B), which form a complex (maturation/mitosis promoting factor, MPF) regulating the G2-M phases of the cell cycle, were also registered. Highly significant differences between HD and ALCL-C were recognized: a) in HD, evidence for abortive mitosis (i.e., difficulty to proceed beyond the metaphase stage) and consequent multinucleation and/or deletion of CD30+ cells was prominent, in contrast to ALCL-C. This was associated with a markedly lower fraction of large atypical cells (LAC) expressing cyclin-B in the cytoplasm and the nucleus (C + N) in HD than in ALCL-C; b) the extent of multinucleation of CD30+ cells in HD, but not in ALCL-C, was correlated with the %p34+ LAC; c) the proportions of LAC expressing p34 and/or cyclin-B (C) were positively related to the percentages of cyclin-B (C + N)+ LAC in ALCL-C but not in HD; d) in HD, in contrast to ALCL-C, the size of the fraction of cyclin-B (C + N)+ LAC did not correlate with the ana/telophase indices (ATI, reflecting successful completion of mitosis) and the magnitude of cell loss; e) in ALCL-C, the percentages of p34+ LAC were positively correlated with ATI or the degree of CD30+ cell deletion, but inversely in HD. With regard to all parameters mentioned above, ALCL-HL tended to take an intermediate position between HD and ALCL-C, but sided more with the latter. In conclusion, our present results suggest a derangement of MPF kinetics and functions that is more profound in HD than in ALCL-C.

Growth patterns of diffuse non-Hodgkin's lymphomas estimated from mitotic and apoptotic indices.

Growth rates of neoplasms could be calculated only on the basis of mitotic and apoptotic indices (MI and AI, respectively), assessed on tissue sections, if the duration of mitosis and apoptosis (Tm and Ta, respectively) in vivo were known. For humans, this is practically never the case. What use then can be made of MI and AI to arrive at a relative, crude estimate of the state of growth? As a model system to study this problem, we chose diffusely growing stage I + II non-Hodgkin's lymphomas (dNHL, n = 94). Cluster analysis revealed the existence of 3 highly distinct groups of dNHL (clusters I, II and III) in the MI vs. AI per case plot, with a roughly linear relation between both parameters. Most nosologic entities defined by the REAL classification comprise cases that were represented in more than one cluster. We adopted the simple formula GI (growth index) = XMI - AI, where X (= Ta/Tm) remains to be evaluated. Based on the assumption that spontaneous regressions of dNHL are rare but do occur, we estimated that X = 2 or, possibly, 3 are best fits for the pooled dNHLs studied. With the assumption of X = 2, (i) 2MI - AI gave relatively lower values for dNHL than proliferative indices such as %Ki-67+ cells; (ii) values for 2MI/AI per cluster showed a pattern inverse to that for %bcl-2+ cells; and (iii) a plot of 2MI - AI vs. 2MI/AI per case allowed the recognition, especially among NHLs with a low cell turnover, of cases where accumulation of presumably longer-lived cells is an important factor in determining growth.

Cell kinetics, morphology, and molecular IgVH gene rearrangements in Hodgkin's disease.

The present study dealt with the question of whether any cellular kinetic patterns correlate with clonal rearrangement of the IgVH gene as revealed by polymerase chain reaction on DNA extracted from lymph nodes with classical Hodgkin's disease (HD) and/or from single CD30+ cells (Hodgkin [H] and Reed-Sternberg [RS] cells). In 15/4 cases with H-RS cells of B or Null phenotype, signs of such monoclonality could be detected (group I) but not in the others (group II). CD30+/H-RS cells in group I differed slightly but significantly from those in group II in that they a) exhibited a larger fraction of cells attaining the anaphase/telophase stage of mitosis, and b) produced relatively more mononucleated cells (H) at the expense of multinucleated (RS) cells. In addition, reactive lymphoid cell (CD30-) infiltrates were considerably less dense in group I that in group II. These findings suggest that the cytokinesis of H-RS cells in group I was moderately more efficient than in group II. However, signs of monoclonality were not associated with the normalization of the mitotic process, which also proved to be disturbed in group I.

Cellular kinetic and phenotypic heterogeneity in and among Burkitt's and Burkitt-like lymphomas.

This study asks whether the known genotypic heterogeneity within and between endemic or sporadic Burkitt's lymphomas (eBLs and sBLs, n = 10 each), and Burkitt-like lymphomas (BLLs, n-12), is reflected in divergent cytokinetics and related immunophenotypes. There was strong evidence that eBL and BLL grow markedly faster than sBL, as shown by differences in mitotic and apoptotic indices. Furthermore, in BLL, the median percentage of neoplastic cells immunoreactive for the bcl-2 protein was much higher than that observed in eBL and sBL. The reverse was true for the median fraction of cells containing c-myc protein. In eBL and sBL, the median fraction of bcl-6 protein-positive cells reached values above 50 per cent, while cells of 8/12 BLLs did not contain detectable amounts of this protein. This observation indicates that in this respect, eBL and sBL resemble normal germinal centres of lymphatic tissue much more than do BLL. Evidence for infection of neoplastic cells by the Epstein-Barr virus (EBV) was observed in 9/10 cases of eBL and in 3/10 of sBL, but not in BLL. EBV-positive lymphomas were associated with distinctly lower apoptotic indices and smaller median percentages of bcl-6-positive cells than EBV-negative tumours.

Mitotic activity and nuclear DNA damage of large cells in Hodgkin's disease: comparison with the expression of p53 and bcl-2 proteins and the presence of Epstein-Barr virus.

The roles of the bcl-2 and p53 proteins in Hodgkin's disease (HD) are poorly understood. We therefore compared their detected presence in Hodgkin and Reed-Sternberg/large atypical (H-RS/LA) cells immunohistochemically with the percentages of these cells double-labeled for CD30 and DNA strand breaks (DNA fragmentation index, DFI); mitotic indices (MI); and the EBV infection status. We found a highly significant inverse correlation between the fractions per case of H-RS/LA cells expressing bcl-2/p53 proteins and the DFI of CD30+ elements. No marked effect of these two oncoproteins on MI was noticed, although these parameters and DFI of CD30+ cells were linearly related. EBV infection of H-RS/LA cells exerted only a limited effect on the parameters tested. The results of this study suggest that overexpressed bcl-2 and, to some extent, p53 proteins in H-RS/LA cells of HD primarily counteract deletion of these cells.

Stage-related differences of mitotic and apoptotic indices, and bcl-2 protein expression in diffusely growing non-Hodgkin's lymphomas.

The present study examined whether growth characteristics of diffusely growing non-Hodgkin's lymphomas (NHL) may differ as a function of stage. Among 105 NHL of various types and sub-types (REAL [Revised European-American Lymphoma] classification), localized (Ann Arbor pathologic stages I + II) lymphomas exhibited clearly higher indices for mitotic activity, apoptosis and cell turnover, as well as a significantly lower percentage of cells containing immunohistochemically detectable bcl-2 protein, than disseminated (stages III + IV) NHL. A similar pattern emerged when high-grade (Kiel classification) lymphomas only were evaluated. Low-grade NHL showed analogous, but less marked, stage-dependent characteristics, with the exception of median percentages of bcl-2+ cells, which remained comparable in all stages. Our findings are consistent with the notion that dissemination of diffusely growing NHL is usually associated with reduced cell turnover and, in high-grade lymphomas, with the generation of longer-lived cells.

Abortive mitoses and nuclear DNA fragmentation in CD30+ large cells of Hodgkin's disease.

This study was undertaken to better comprehend the reasons for the scarcity of Hodgkin and Reed-Sternberg (H-RS) cells in Hodgkin's disease (HD) despite their expression of "proliferation-associated antigens". To this end, we assessed the relative frequency of mitotic phases and nuclear damage (detected by in situ end-labeling of DNA strand breaks) in CD30+ large cells of nodular sclerosis and mixed cellularity HD. Our results show that a) most CD30+ cells in HD exhibit abortive mitoses, with a highly significant arrest at the metaphase-ana/telophase transition, and b) many of these elements, i.e. mainly H-RS cells, show fragmentation of nuclear DNA, suggesting imminent or actual death. Percentages of CD30+ cells that entered mitosis and those with DNA strand breaks were of a similar order of magnitude and correlated significantly in a linear fashion. These findings are consistent with the concept that cell deletion is the major cause of the paucity of H-RS cells in HD.

Growth vs. DNA strand breaks in Hodgkin's disease: impaired proliferative ability of Hodgkin and Reed-Sternberg cells.

We re-appraised the cell renewal pattern in Hodgkin's disease (HD), considering that most, though not all, Hodgkin/Reed-Sternberg (H-RS) cells exhibit abortive mitoses and that a substantial fraction of these exhibits DNA damage suggestive of imminent or actual cell death. Using combined immunohistochemistry and in situ end-labeling to detect strand breaks, the percentage per case of CD30+ (mainly H-RS) cells with DNA fragmentation (DNA fragmentation index [DFI]) was estimated. For each case, we registered the mitotic index (MI) of CD30+ cells and the percentage of Ki-67+ atypical large cells. To quantify the sum of our parameters for mitosis, whether successful or not, and DNA damage, we introduced the kinetic event index (KEI = MI + DFI). Only DFI and KEI distinguished significantly between mixed cellularity and nodular sclerosis HD. The values for MI and DFI, and therefore for KEI, CD30+ and CD30- small lymphoid cells were proportional. The percentages of Ki-67+ large atypical cells (median 50%) did not correlate significantly with either MIs or DFIs of CD30+ cells. Cluster analysis revealed the existence, independent of histological subtype, of 2 large groups of HD with different KEIs. Our findings suggest that cell deletion plays an important role in HD. Further, it appears that proliferation-associated antigens in H-RS cells do not reflect successful cell production in this disorder.

Novel, contrast gradient-oriented, automated chromatin texture analysis. I. Feasibility study on nuclei from benign and malignant breast epithelial cell lines in fine needle aspirates.

Coarse granularity of nuclear chromatin texture is a prominent feature of most malignant cell lines. We have chosen the abrupt transition from eu- to heterochromatic foci (high contrast gradient [CG]) as a novel parameter for coarseness. This feature was quantified using automated image analysis of single nuclei in smears stained by the May-Grünwald-Giemsa technique. The principle of this approach consists of eliminating, with the help of subtraction between two image lowpass filters, the small grey level differences among pixels, so that only high CG values are retained on the digitized image. The sum of these distinctive microareas is then taken as a fraction of the area of the peripherally eroded nucleus, and this ratio is designated as contrast gradient index (CGI) per nucleus. This method was tested on fine needle aspirates from 11 patients with benign breast disease (BBD) and 14 with mammary carcinoma (CA). For each specimen, 60 nuclei were analyzed, with a measuring time per nucleus of about 1 min. A high significant distinction between epithelial cell populations in BBD and CA, respectively, was obtained by variance analysis of all CGIs per nucleus (p = 2 x 10(-18). The median and the mean values of CGI per specimen were the next best discriminators, followed by the modes and the standard deviation of CGI per specimen. The percentage of nuclei per specimen with CGI values of greater than 12 was also significantly greater in CA than in BBD.

Diffuse centrocytic and/or centroblastic malignant non-Hodgkin's lymphomas: comparison of mitotic and pyknotic (apoptotic) indices.

Mitotic indices (MIs) and pyknotic (apoptotic) indices (PIs) were assessed in diffuse centrocytic (CC, n = 10), centroblastic/centrocytic (CB/CC, n = 18) and centroblastic (CB, n = 20) malignant non-Hodgkin's lymphomas (NHL). Significant differences were observed. MIs were lowest in CC (median: 0.07%), intermediate in CB/CC (0.18%) and highest in CB (0.43%) NHLs. The PIs exhibited a similar pattern. The PIs of CC (0.11%) and CB/CC (0.17%) NHLs were significantly different from those of CB lymphoma (0.62%). The ratios MI/PI per case, as well as MIs and PIs per case, varied greatly and showed considerable overlapping, thus documenting a marked inter-case and inter-group heterogeneity. MIs tended to loosely correlate with PIs in a non-linear fashion, which raises the question of feedback mechanisms. More information is needed on mitotic time (TM) and apoptotic time (TA), in order to estimate cell doubling time from data on MIs and PIs.

Hodgkin's disease and CD30-positive anaplastic large cell lymphomas--a continuous spectrum of malignant disorders. A quantitative morphometric and immunohistologic study.

The authors have examined cellular areas of lymphoma tissue in 28 cases of Hodgkin's disease (HD) or anaplastic large cell lymphoma (ALCL, 'Ki-1 cell lymphoma') to evaluate the boundaries between the two entities. Methods applied included conventional histology; test point analysis; semiautomated morphometry of nuclear profile features of Reed-Sternberg and other atypical large cells (RSALCs); and immunohistochemistry of these elements on all paraffin sections and, in 15 cases, on frozen sections. Mean nuclear profile morphotypes of RSALCs per case varied independently of immunophenotype and histologic diagnosis. Conversely, immunohistochemistry demonstrated significant, although not consistent, preferential positivities of these CD30+ elements for CD15 in HD, and for epithelial membrane antigen (EMA) and CD43 in ALCLs. In the latter, RSALCs also exhibited a tendency for CD45 and CD45RO positivity and for the expression of T-cell-associated antigens. However, there were considerable overlaps. This continuous spectrum of RSALC nuclear profile morphotypes and immunophenotypes, ranging from HD over questionable cases, intermediate between HD and ALCL, to ALCLs, was paralleled by differences in the reactive component of lymphomas. Lymphocytes and granulocytes were significantly deficient in ALCLs.