R. Brent Archibald and John H. Frenster,
Division of Oncology, Department of Medicine,
Stanford University School of Medicine, Stanford, California 94305
Lymph nodes from 5 untreated patients with the nodular-sclerosis form of Hodgkin's disease were analyzed ultrastructurally for the cellular interaction between lymphocytes and neoplastic cells occurring in-vivo. A direct and significant correlation was observed between the tightness of lymphocyte apposition and the ultrastructural cytotoxic changes within neoplastic cells, which supports the concept that lymphocyte immunity against neoplastic cells may be important in the pathogenesis and prognosis of Hodgkin's disease in untreated patients.
Pathologists have long noted the infiltration of lymphocytes into human neoplasms and more recently have correlated the degree of such infiltration with the clinical course and prognosis of the neoplasm for individual patients (1). Among such human neoplasms , Hodgkin's disease demonstates a direct correlation between the degree of lymphocyte infiltration and patient survival (2). In addition, both the peripheral blood lymphocyte count (3) and the degree of reactivity of blood lymphocytes to defined mitogens (4) reflect the clinical stage and prognosis of individual patients.
The recognition that many human neoplasms possess tumor-associated antigens (5), to which the patient responds with lymphocyte-mediated immunity (6), has suggested that the infiltration of lymphocytes into untreated neoplasms may represent an immune response by the patient to the neoplasm, analogous to a homograft reaction (1). The recent recognition that tumor-associated antigens may be present in Hodgkin's disease (7) has increased the likelihood that lymphocytes may be mediating an immune response against the neoplasm in untreated patients with Hodgkin's disease (8, 9).
Lymphocyte immunity against target cells is manifested in-vitro as a close apposition of the lymphocytes against the target cells before the target cells are destroyed (10) by mechanisms which may utilize lymphocyte plasma-membrane adenosine triphosphatase (11). Previous electron microscopic studies of Hodgkin's disease (12, 13) did not use fixation techniques which would have permitted an analysis of interactions between lymphocytes and neoplastic cells; in fact, those studies were largely confined to the ultrastructural analysis of individual Reed-Sternberg and Hodgkin's cells (12, 13). We recently developed (9, 14) quantitative methods for an ultrastructural analysis of cell interactions in lymph nodes more adequately fixed for the preservation of in-vivo structure than were those previously available (12, 13).
All 5 study patients sampled were previously untreated and had a histologic diagnosis of Hodgkin's disease, nodular-slerosis (NS) type, made in the Division of Surgical Pathology. All patients had Stage III disease following complete pathologic staging (15), all were free of constitutional symptoms, and all had peripheral blood lymphocyte counts > 1500/mm3. Biopsies of lymph nodes in such patients were made for diagnostic staging, and those selected for the current study were confined to lymph nodes not previously seen by lymphangiography (16). All lymph nodes were subsequently confirmed as positive for Hodgkin's disease, NS type, by concurrent thick-section light microscopy (16).
Random samples of each lymph node were placed, within 30 seconds of excision by the surgeon, into ice-cold Eagle's minimal essential medium (Gibco), at pH 7.4, containg 5% glutaraldehyde, and were then brought to the laboratory where they were sectioned into 1 mm3 cubes. These cubes were then postfixed in 1% OsO4, dehydrated in ethanol, embedded in Epon, sectioned at 0.1 um thickness, stained with 5% uranyl acetate, and examined at 80 kv under high resolution in a Siemens 1A electron microscope as previously described (17).
The ultrastructural classification of Bernhard and Leplus (12) was used to identify neoplastic polynuclear Reed-Sternberg cells, mononuclear Reed-Sternberg cells, and mononuclear Hodgkin's cells (12). Each of these neoplastic cells was coded and imaged on a projector plate at 10,000 magnification. Five prints of each cell were then prepared with an additional four-fold magnification. One print was employed by an independent observer to identify the neoplastic cell type. A second print was used by an independent observer to identify all non-neoplastic cells. A third print was utilized for a quantitative sector analysis (9, 14) of each neoplastic cell (outlined in black ink); a Lietz map reader was used to yield neoplastic cell circumference values in mm. Each surrounding cell apposing the neoplastic cell was assigned a sector of the neoplastic cell circumference. This sector was measured in turn by the map reader; the measured sector length was then expressed as a decimal fraction of the total neoplastic cell circumference (9, 14). The surrounding apposed non-neoplastic cells were identified as either lymphocytes, macrophages, fibroblasts, plasma cells, eosinophiles, neutrophils, or endothelial cells (18) and were then identified by an independent observer as in either tight or loose apposition to the neoplastic cell (9, 14). The sectors of lymphocytes in tight apposition to a single neoplastic cell were then summed as a total decimal fraction of the neoplastic cell circumference for each neoplastic cell. A fourth print was utilized by an independent observer to estimate the degree of ultrastructural cytotoxic change within each neoplastic cell, with special attention given to mitochondrial swelling, nuclear edema, disruption of the endoplasmic reticulum, and irregularities of the plasma membrane (19). Finally, the degree of such neoplastic cell cytotoxicity was correlated with the decimal fraction of the cell circumference occupied by the lymphocytes in tight apposition to the particular neoplastic cell.
Random sampling of lymph nodes from untreated patients with Hodgkin's
disease, NS type, revealed all cell types as originally described by Bernhard
and Leplus (12).
Figure 1A (left, above). Polynuclear Reed-Sternberg cell (1) surrounded
by small lymphocytes (2), a macrophage (3), and a fibroblast (4). Neoplastic
cell shows no significant ultrastructural cytotoxic change, and surrounding
cells are all in loose apposition. Neoplastic cell (outlined in black ink)
and sectors of surrounding cells numbered from 1 to 32. (X 2,000).
Figure 1b (right, above). Mononuclear Reed-Sternberg cell (1) with deeply hyperlobated nucleus, hypertrophic nucleoli, and surrounding lymphocytes (2), some of which orient themselves to the circumference of the neoplastic cell. Neoplastic cell shows a minimal degree of cytotoxic change (slight nuclear edema); surrounding lymphocytes are in moderately tight apposition. (X 3,000).
Figure 1D (right, above). Mitotic neoplastic cell, in late telophase with complete reconstitution of the nuclear membranes (pentalaminar) of both daughter nuclei (1 and 1), but with no indication of cytoplasmic division, suggesting the formation of a binucleated cell. Surrounding medium-sized lymphocytes (2 and 2) are in loose apposition. (X 4000).
Lymphocytes comprised approximately 75% of all cells in the sampled lymph nodes and were either of the small or medium-sized type (20), with single regular nuclei, cytoplasmic monosomes, scanty endoplasmic reticulum, and no evidence of nuclear activation (21). Macrophages, fibroblasts, plasma cells, eosinophiles, and endothelial cells, as observed in sampled lymph nodes, will be described elsewhere (22). Lymphocytes were frequently found in parallel orientation to the surface of neoplastic cells (Figs. 1A and 1B), and sometimes were seen deeply invading those neoplastic cells displaying moderate degrees of cytotoxic change (Fig. 1C). Uropod formations on lymphocytes were occasionally observed to be directed toward neoplastic cells.
The first 25 neoplastic cells collected by random sampling in equal
numbers from each of the 5 study patients in each of the categories "no,
minimal, or moderate degrees of ultrastructural change", respectively,
were correlated with the decimal fraction of their total cell circumference
occupied by lymphocytes in tight apposition.
Text-Figure 1. Relationship of tight lymphocyte apposition to ultrastructural
cytotoxic change occurring within 25 neoplastic cells in each of the categories
of no, minimal, or moderate degrees of such toxic change.
By use of rapid fixation in cold solutions of glutaraldehyde in the operating room, the quality of electron micrographs of sampled lymph nodes from untreated patients with Hodgkin's disease has improved, compared to previous studies in which only osmic acid fixation was used (12, 13). This improvement has permitted the analysis of lymphocyte and macrophage apposition to neoplastic cells in Hodgkin's disease (9, 14) and has revealed a significant correlation between tight apposition by lymphocytes to neoplastic cells and ultrastructural cytotoxic changes within such neoplastic cells (Text-Figure 1).
It is not yet apparent whether lymphocyte tight apposition is an independent or dependent variable in the cellular interaction, or whether such apposition plays a causal role in the observed ultrastructural cytotoxicity within the neoplastic cells. However, within in-vitro systems, tight apposition by immune lymphocytes is a prerequisite for the expression of cytotoxicity within target cells to which the lymphocytes have been sensitized (10). Such cytotoxicity may be mediated via the increased levels of lymphocyte plasma membrane adenosine triphosphatase, which is found in immune lymphocytes when they are exposed to target cells to which they are sensitized (11).
Lymphocytes in the peripheral blood of untreated patients with Hodgkin's
disease often are larger and more active in DNA synthesis than normal (8),
but late in the course of the disease they often decrease in mumber (3)
and become less responsive to defined mitogens such as phytohemagglutinin
(4). Advanced neoplasms in man are associated with increased levels of
free-circulating tumor antigen (23). Such free antigen in the presence
of small quantities of antibody can decrease the activity of lymphocytes
specifically sensitized to the antigen (24) and may account for the loss
of lymphocyte reactivity in advanced Hodgkin's disease (3, 4). Decreases
in peripheral blood lymphocyte counts in advanced Hodgkin's disease may
relate to direct invasion by neoplastic cells of such vital lymphopoietic
organs as the thymus and the spleen. These relationships between neoplastic
cell proliferation, increasing quantities of tumor-associated antigen,
and lymphocyte immunity and inactivation are summarized in Text-Figure
2,
Text-Figure 2. Possible lymphocyte - Reed-Sternberg cell interactions
in Hodgkin's disease. Neoplastic cells may possess a tumor-associated antigen
(see text) which can sensitize thymus-dependent T-lymphocytes. Such sensitized
lymphocytes, when activated, can antagonize the proliferation of the neoplastic
cells by inducing cytotoxic changes within such cells (see text). Should
immunosuppresion of the lymphocytes or antigen-deletion of the neoplastic
cells occur, neoplastic cell proliferation could progress, with the production
of excessive quantities of soluble tumor antigen and the resultant inactivation
of sensitized lymphocytes (see text). Neoplastic cell proliferation can
also decrease T-lymphocyte counts by the direct invasion of such lymphopoietic
organs as the thymus and the spleen (see text).
Confirmation of the role of antigen, antibody, and lymphocytes in the pathogenesis and prognosis of Hodgkin's disease would prompt consideration of such therapeutic manipulations as non-specific T-lymphocyte activation (1), plasmapheresis of circulating antigens, and immuno-suppression of B-lymphocytes as a source of inactivating antibody (1).
Supported in part by Public Health Service grants CA10174 from the National Cancer Institute and AM01006 from the National Institute of Arthritis and Metabolic Diseases; by grant IC-45 from the American Cancer Society; and by a Research Scholar Award from the Leukemia Society to Dr. Frenster.
1. Frenster JH, and Rogoway WM, "Immunotherapy of Human Neoplasms with Autologous Lymphocytes Activated In-Vitro", in: "Proceedings of the Fifth Annual Leukocyte Culture Conference". (Harris J, ed.), pp. 359-373, (1970), New York: Academic Press.
2. Keller AR, Kaplan HS, Lukes RJ, et al, "Correlation of Histopathology with Other Prognostic Indicators in Hodgkin's Disease", Cancer 22: 487-499 (1968).
3. Young RC, Corder MP, Haynes HA, et al, "Delayed Hypersensitivity in Hodgkin's Disease", Am. J. Med. 52: 63-72 (1972).
4. Han T, Sokal JE, "Lymphocyte Response to Phytohemagglutinin in Hodgkin's Disease", Am. J. Med. 48: 728-734 (1970).
5. Gold P, "Antigenic Reversion in Human Cancer", Ann. Rev. Med. 22: 85-94 (1971).
6. Hellstrom I, Hellstrom KE, Sjogren HO, et al, "Demonstration of Cell-Mediated to Human Neoplasms of Various Histologic Types", Int. J. Cancer 7: 1-16 (1971).
7. Order SE, Porter M, and Hellman S, "Hodgkin's Disease: Evidence for a Tumor-Associated Antigen", New Eng. J. Med. 285: 471-471 (1971).
8. Crowther D, Fairley GH, and Sewell RL, "Lymphoid Cells in Hodgkin's Disease", Nature 215: 1086-1088 (1967).
9. Archibald RB, and Frenster JH, "Interactions of Lymphocytes with Reed-Sternberg Cells within Human Hodgkin's Disease Lymph Nodes", J. Cell Biol. 47: 8a-9a (1970).
10. Biberfeld P, Holm G, and Perlmann P, "Morphological Observations on Lymphocyte Peripolesis and Cytotoxic Action In-Vitro", Exp. Cell Res. 52: 672-677 (1968).
11. Dimitrov NV, and Ellegaard J, "Elevated Lymphocyte Adenosine Triphosphatase Activity in Patients with Gastrointestinal Carcinoma", New Eng. J. Med. 286: 353-355 (1972).
12. Bernhard W, and Leplus R, "Fine Structure of the Normal and Malignant Lymph Node", New York: Macmillan & Co. (1964).
13. Mori Y, and Lennart K, "Electron Microscopic Atlas of Lymph Node Cytology and Pathology", New York: Springer-Verlag, (1969).
14. Archibald RB, and Frenster JH, "Quantitative Electron Microscopic Sector Analysis of Lymphocyte Interactions with Reed-Sternberg Cells in Hodgkin's Disease", Proc. Am. Assoc. Cancer Res. 12: 43-44 (1971).
15. Glatstein E, Guernsey JM, Rosenberg SA, et al, "The Value of Laparotomy and Splenectomy in the Staging of Hodgkin's Disease", Cancer 27: 1277-1294 (1969).
16. Kadin ME, Glatstein E, and Dorfman RF, "Clinico-Pathologic Studies of 117 Untreated Patients Subjected to Laparotomy for the Staging of Hodgkin's Disease", Cancer 27: 1277-1294 (1971).
17. Frenster JH, "Electron Microscopic Localization of Acridine Orange Binding to DNA within Human Leukemic Bone Marrow Cells", Cancer Res. 32: 1128-1133 (1971).
18. Lukes RJ, and Butler JJ, "The Pathology and Nomenclature of Hodgkin's Disease", Cancer Res. 26: 1063-1083 (1966).
19. Fawcett DW, "The Cell, its Organelles and Inclusions. An Atlas of Fine Structure", Philadelphia: W. B. Saunders Co. (1966).
20. Anderson DR, "Ultrastructure of Normal and Leukemic Leukocytes in Human Peripheral Blood", J. Ultrastruct. Res. Suppl. 9: 1-42 (1966).
21. Tokuyasu K, Madden SC, and Zeldis LJ, "Fine Structural Alterations of Interphase Nuclei of Lymphocytes Stimulated to Growth Activity In-Vitro", J. Cell Biol. 39: 630-660 (1968).
22. Frenster JH, and Archibald RB, "Electron Microscopy of In-Vivo Macrophage Activity Against Human Lymphoma Cells", In Preparation.
23. Thomson DM, Krupey J, Freedman SO, et al, "The Radio-Immunoassay of Circulating Carcinoembryonic Antigen of the Human Digestive System", Proc. Natl. Acad. Sci. U.S., 64: 161-167 (1969).
24. Sjogren HO, Hellstrom I, Borsal SC, et al, "Suggestive Evidence that the Blocking Antibodies of Tumor-Bearing Individuals May Be Antigen-Antibody Complexes", Proc. natl. Acad. Sci. 68: 1372-1375 (1971).
25. Order SE, and Hellman S, "Pathogenesis of Hodgkin's Disease", Lancet 1: 571-573 (1972).
26. Jehn UW, Nathanson L, Schwartz RS, et al, "In-Vitro Lymphocyte Stimulation by Soluble Antigen from Malignant Melanoma", New Eng. J. Med. 283: 329-333 (1970).
0. Electron Microscopy of Human Lymphocytes before and after Activation by PHA (Busch H, 1974).
1. FrensterJH, Papalian MM, Masek MA, and Frenster JA, "Electron Microscopic Analysis of Lymph Node Cellular Activity in Hodgkin's Disease", J. Natl. Cancer Inst. 63: 331-335 (1979).
2. Frenster JH, "Single Cell Analysis of DNase I-Sensitive Sites during Neoplastic Cell Differentiation within Hodgkin's Disease Lymph Nodes", in: "Leukemia Reviews International", (Rich MA, ed,.), Vol. 1, pp. 22-23 (1983), Marcel Dekker, Inc., New York.
3. Hodgkin's Disease Abstracts Published, November, 1998.
4. Hummel M, Marafioti T, and Stein H, "Clonality of Reed-Sternberg Cells in Hodgkin's Disease", New Eng. J. Med. 340: 394-395 (1999).