- Email: [email protected]
Nuclear morphologic and morphometric analyses of large noncleaved cell and immunoblastic non-Hodgkin's lymphomas
Nuclear Morphologic and Morphometric Analyses of Large Noncleaved Cell and Immunoblastic Non-Hodgkin's Lymphomas IRVING DARDICK, MD, FRCP(C], DOUGLAS R. CALDWELL, BS, RT, W. T. ELLIOTr McCAUGHEY, MD, FRCP{C], AND MAHA AL-JABI, MD, FRCP[C] systematic study of nuclear features has been undertaken in mycosis fungoides. 25-33 Many aspects o f the Lukes-Collins system o f N H L classification are based on a hypothetic seq u e n c e o f n o r m a l l y m p h o c y t e t r a n s f o r m a t i o n in which changing nuclear structure is a central feature. 9-11 However, recent investigations of lymphocyte transformation in h u m a n peripheral T cells 34 and mouse spleen B cells35 did not confirm the sequence of small and large, cleaved and noncleaved nucleated cells envisaged in this concept. Instead,'a progressive increase in nuclear size is coupled with a gradual disaggregation o f c o n d e n s e d c h r o m a t i n clumps. T h e morphologic aspects of current NHL classifications, such as nuclear size, configuration, and condensed chromatin organization, require detailed analyses by objective techniques. Such studies are needed to test the adequacy and degrees of reliability of current morphologic criteria for segregating NHL into the variety of subtypes used for classification, therapeutic, and prognostic purposes and to establish what, if any, relation exists between tile various subcategories and normal lymphocytes. Two recent clinicopathologic studies o f the morphologic subcategories of large cell lymphoma illustrate the inadequacy of current methods in the histologic evaluation of NHL. 4,15 The investigation in this report is part of a systematic evaluation of nuclear features in the subtypes of NHL, segregated by morphologic criteria. T h e results reported are summary data derived from nuclear morphometric image analysis of a series composed of large noncleaved cell lymphomas and immunoblastic sarcomas, and o f lymphocytes f r o m various regions of lymph nodes involved by reactive hyperplasia. T h e results reveal reasons for interobserver discrepancies in histologic diagnoses and inconsistencies in the diagnostic parameters of current NHL classifications.
Both morphologically and immunologically, non.Hodgkin's lymphoma (NHL) of the large cell type has been shown to be a heterogeneous category. However, the homogeneity of the nuclear p a r a m e t e r s , particularly size and condensed chromatin organization, used to classify this subtype of NHL has not been investigated. In fact, objective morphologic techniques have not been systematically applied to verify the segregation of NttL on the basis of nuclear parameters, a concept common to all current classification systems. In this study morphometric image analysis was used to compare the nuclei in 20 specimens from NHLs of the large cell type with.those in mantle zone and germinal center lymphocytes from lymph nodes with reactive hyperplasia. R e s u l t s of the assessment of mean nuclear area in large cell lymphomas revealed that this class is also heterogeneous, with s o m e of the specimens having a nuclear size in the u p p e r range of that for normal small lymphocytes. In addition, in only a few of these specimens was the mean nuclear area within the range of that for fully transformed germinal center lymphocytes. The majority of large cell lymphomas have a nuclear size more characteristic of partially transformed lymphocytes in germinal centers. In addition to indicating inconsistencies in the current diagnostic criteria used in NHL classifications, the results indicate r e a s o n s for interobserver variations in clinicopathologic trials; the validity of nuclear size as a prognostic indicator and the biologic basis for classifying NHL as a reflection of normal lymphocyte transformation are also questioned. In terms of patient management, the classifications of NHL currently used require objective reappraisal. HUM PATHOL 15:965--972, 1984.
Recurring problems in clinical trials and pathologic studies o f n o n - H o d g k i n ' s l y m p h o m a (NHL) have been the lack of reproducibility and the subjective bias inherent in such studies. 1-5 Ahhough immunologic, morphologic, and prognostic studies have confirmed the heterogeneity of the large cell class of NHL, 6-16 difficulties in standardizing the specific morphologic criteria and definitions for each subcategory of large cell lymphoma persist. 4,6 One of the major reasons for the inconsistencies in morphologic diagnoses in NHL has been the reliance on subjective evaluation o f the cytologic features of neoplastic lymphocytes. Despite the important role of nuclear morphology in current classification systems, this organelle has been evaluated objectively only recently in some categories of NHL. 17-24 A more
MATERIALSAND METHODS
Material for this study was selected from cases that had been accessioned at the Canadian T u m o u r Reference CEntre, Ottawa, Canada, with diagnoses of diffuse or nodular histiocytic lymphoma and in which wet tissue or plastic resin-embedded tissue was available. Cases were examined independently by three of the authors (I.D., W.T.E.Mc. M.A.-J.), and in each a
Received October 7, 1983, from the Canadian Tumour Reference Centre, Ottawa, Ontario, Canada. Accepted for publication N o v e m b e r 17, 1983. Supported by a grant from the National Cancer Institute of Canada. Address correspondence and reprint requests to Dr. Dardick: Canadian T u m o u r Referenc.e Centre, Clinical Studies Unit Building, 60 Ruskin Avenue, Ottawa, Ontario, Canada KIY 4M9.
965
HUMAN PATHOLOGY
Volume 15, No. 10 [October 1984)
consensus diagnosis was established. By criteria established in the literature, 9-11,36-38 neoplasms composed primarily of large or medium-sized lymphoreticular cells that had the characteristics and cytologic details of large noncleaved cell lymphoma or immunoblastic sarcoma were segregated for morphometric analysis. Routinely p r e p a r e d histologic sections showed these tumors to be composed basically of neoplastic lymphocytes with generally smoothly contoured vesicular nuclei; one to three acidophilic nucleoli; and multiple, small, diffusely a r r a n g e d clumps of condensed chromatin. Tumors that had any significant component (>20 per cent) of small cleaved cells, a n d that might thus be classified as mixed, were excluded. Of the 20 tumors measured, 16 were diffuse large noncleaved cell lymphomas, two were nodular'large noncleaved cell lymphomas, and two were immunoblastic sarcomas. For comparative purposes, six lymph node biopsy specimens with a diagnosis of reactive hyperplasia and in which wet tissue was available were also selected for morphometric analysis. Histiocyte nuclear parameters were measured in prominent aggregates of this cell type in a lymph node involved.with atypical lymphoid hyperplasia. For both the lymphomas and the lymph nodes with reactive hyperplasia, the majority of tissues had been fixed in 10 per cent buffered formalin. One large cell lymphoma had been fixed in B-5, while seven specimens had been fixed initially in Karnovsky's solution. Wet tissue was processed for glycol . methacrylate resin e m b e d d i n g (Sorvall e m b e d d i n g medium, D u p o n t Instruments, Wilmington, Delaware), and sections were cut 1 p~m thick with a Sorvall JB4 microtome with glass knives and stained with celestine blue 39 or Gill's hematoxylin (Fisher Scientific, Pittsburgh, Pennsylvania) and eosin. Tissues fixed in Karnovsky's solution had been embedded previously in E p o n - A r a l d i t e resin; with a LKB IV ultramicrotome, 1-1~m sections were cut and stained with tolu{dine blue. For morphometric analysis, random but representative regions of the tumors and representative areas o f the mantle zoiae and germinal centers o f reactive lymph nodes were p h o t o g r a p h e d in nonoverlapping frames on 35-mm film with an oil immersion objective (100 x ). Each frame of the negative film was projected onto a digitizer pad (Summagraphics Corporation, Fairfield, Connecticut) at a final magnification of 2500 x . In each frame all welloutlined lymphocyte nuclear profiles were traced, and the data were collected and analyzed with a MSA II program (Atlantis Scientific Systems Group, Ottawa, Canada) and an Apple II Plus microcomputer (Apple Computer, Cupertino, California). The digitized information was used to calculate area, perimeter, volume, and contour index of each nuclear profile. For profiles having well-defined angulated or narrow identations, the depth of this identation was measured from the margins of the nuclear membrane on either side of the invagination to its base.
Contour index 4~ is a size-independent measurement of the shape of a profile in which increasing irregularity or eccentricity o f a profile results in ihigher values (a circle has a contour index of 3.54). T h e chromatin organization of normal and neoplastic nuclei was assessed by descriptive criteria reported previously. 34,35 Essentially, lymphocyte nuclei are subdivided into three types (fig. 1, left): morphotype 1 involves the prominent and compact masses of c o n d e n s e d c h r o m a t i n seen in u n s t i m u l a t e d small lymphocytes; morphotype 3 demonstrates the markedly disaggregated condensed chromatin and prominent nucleolus characteristic of the large, fully transformed lymphocyte; and morphotype 2 is characterized by nuclei with numerous, small to moderately sized clumps of condensed chromatin and thus represents the intermediate degree of condensed chromatin disaggregation a p p a r e n t in partially transformed lymphocytes.34,35 RESULTS
Each data point in figures 2 to 5 represents the mean value for a particular nuclear parameter measured morphometrically in each sample of lymphoid tissue. In the six normal but reactive lymph node biopsy Specimens, a total of 11,618 lymphocyte nuclei were measured: 4,160 m o r p h o t y p e 1 mantle zone (M1) l y m p h o c y t e nuclei (mean per sample, 693; range, 426 to 1,051); 1,800 morphotype 1 germinal center (GC1) lymphocyte nuclei (mean per sample, 300; range, 194 to 434); 4,381 morphotype 2 germinal center (GC2) nuclei (mean per sample, 730; range, 501 to 1,106); and 1,277 morphotype 3 germinal center (GC3) nuclei (mean per sample, 212; range', 107 to 519). Nuclear profiles traced in the 20 samples from large cell lymphomas totalled 41,020. Of these, 7,122 were categorized as m o r p h o t y p e 1 (LC1) nuclei (mean per sample, 356; range, 41 to 732); 12,544 as morphotype 2 (LC2) (mean per sample, 672; range, 335 to 997); and 21,354 as m o r p h o t y p e 3 (LC3) (mean per sample, 1,068; range, 521 to 1,708). Nuclear Profile A r e a
For comparative purposes, figure 2 illustrates tile ranges o f mean nuclear profile areas for normal, small-unstimulated lymphocytes (morphotype 1) in mantle zones and germinal centers, and for partially transformed (morphotype 2) and fully transformed (morphotype 3) lymphocytes of germinal centers. For typical neoplastic lymphocytes in large cell lymphomas (LC2 and LC3), mean nuclear size ranged from 12 to 50 p.m 2. Even morphotype 3 lymphocytes (LC3), with their open, vesicular nuclei, had a nuclear size range of 18 to 50 ~m~ (fig. 2). Unexpectedly, the mean nuclear profile areas for small morphotype 1 lymphocytes in the large cell lymphomas did not fall within tile range of morphotype 1 nuclei for normal tissues, but were generally smaller. Similarly, the majority of mean nuclear areas for morphotype 3 tumor 966
NUCLEAR ANALYSESIN NHL CDardick et al.l
k~,.';,.-r
+" 2 : - - ", ' + ~ r +4:'~+;
)
"
.:'
"~-',"
:
~,~-.~
,
-
;+
-
,
.
.,-~ "---"
"
~ .
",-'')2 "-,+
~:~:~
,
,~.:;4
,~" ,r . . . .
+
+.
9
l+
",r~)+,
,:.
s;~
"~
~-./.
,+
,.;
" .:
*.% 9
"
'
"-':
.
;"
C.'":.':~-._)'.~ .. -,,.
.
. ~ . , . ~ ,
:~.
.... ..
-.,
.-~
"
O" _a,_
-
.J
~:-~.
. -,~.
+..,,
.+,
'~t.'i..:
~
+,, ,
;': "
-~'~ll~'..-'_-Y~".v,
"
,LJ",
r
"~'.
". '
:~
.
-+-\7
~
C-,.
~.,-~
,.r ;._..
, I~ + . , + 3 ' 9 .. . ,.+.7:> ~_:1--~ +-+-* .': 9 "-+'.+-~+_~,. ".~ ---:++" . " . .' . . " "'r, .... ~" : ' " " * " "'~" -+','"+"~.."
'
...
~ - - ..
~.+,';+'
.
iai~
.
."~'~..'~ ",%~+'+~5.."
'-:
g.,j
,~---. ~ i , ~ . ' - ~ , , I, ~ 3 " " ... ,. L.++'U;" [ +i. _2 *'~/~-~ID . 9 ", + " . . ~. ;. , .
.
~ ' ~
+
. .,,,
9
- +" " .
. . . .
'~.
~
.""~ "+<'~ ~
!1~
.
,.,,.'i~It
?r,-L.,...-+.),
.._:, . .
"~'"" +,',,.+, .'~k.+'r"
-,
. [~..'. 9" ' ;
t~ 2. ,)
-'--~'~ ..
,-v .-.
--':
TM
e_~
i,,,,
f.,.. : p.+j
-~ .+
9
. . . . .
).
- . J
"
, ,'.
~r~.," 9 - ,
, ~:.'=-, (-~" . . . .
f~.,
(_~
~
..'v.
~# '" ~ - ' -
-,---.~-~
"
e~..
,"
"t-+---
-++'++., "+ b
v.--~:
"-,,
9 "+,
; , /-.~
"
/~;-".-~
~;
~-
~.-~--,,I~. -+ ~ ' ~
_-5--.~
......
I
.
,+.,"
' '
9 -
+..=',
, . :-,.. --:.-'~: , - ' . J_'f,Z T;~',-.': , . , + , ~ 2 < 3 . :2" : - . ' - " , , . "_ c-~,w.._~+.~_.: +.i; , - - z ~ ~'~" _'~; FIGURE 1. Large cell lymphoma, Epon-embedded sections Left, diffuse lymphoma from case 1. Right, nodular lymphoma from case 2. Examination with a x 40 objective of l-i~m sections clarifies the mean nuclear area differences recorded for the two cases presented in table I. Numerals designate typical examples of the three nuclear morphotypes. (Toluidine blue stain, x 550.)
cells in large cell lymphomas were less than the lowest mean nuclear area values for morphotype 3 lymphocytes in normal germinal centers (fig. 2). In fact, a n u m b e r of the large cell lymphomas had mean nuclear areas within the range for morphotype 1 lymphocytes of the mantle zone and germinal center. Although varying in condensed chromatin distribution, m o r p h o t y p e 2 and 3 nuclei in the large cell lymphomas had similar ranges of mean profile area. It is evident that mean nuclear area values for formalinand Karnovsky's solution-fixed t u m o r tissues were similar (fig. 2). Figures 1 (left) and 6 (le/t) represent regions from one of the large cell lymphomas in which the areas of neoplastic lymphocyte nuclei were in the upper range, while figures 10ight) and 6 (right) demonstrate the lower range of mean nuclear area (fig. 2 and table 1). Figure 6 shows paraffin-embedded tissue photographed with a x 2 5 objective, and figure 1 shows E p o n - A r a l d i t e plastic-embedded sections photographed with a x 40 objective. In the former it was difficult to appreciate differences between the two examples in nuclear area; the differences were more readily apparent in the plastic-embedded tissues. In both large cell lymphomas (fig. 6, left and right), it was easier to detect differences in nuclear size between the lymphocytes with small, darkly staining nuclei (LC1) and the obvious neoplastic nuclei (LC2 and LC3). T h e similarity of the mean nuclear areas of LCl-type cells in these two'large cell lymphomas and
the differences between tile LC1 population of lymo phocytes and the LC2- and LC3-type cells in case 2 are evident in table 1.
Nuclear Contour Index [NCI) A narrow range of NCI, and only minimal deviation from a circular outline, characterized the nuclei of normal lymphocytes, whether in the mantle zone or germinal centers (fig. 3). All three nuclear subtypes in large cell lymphomas also had similar ranges o f m e a n NCI. However, unlike those in n o r m a l lymphocytes, n u c l e a r profiles in the lymphomas were generally considerably more irregular, as reflected by the significantly higher mean NCI of the majority of specimens (fig. 3). Ahhough the degrees of-irregularity and the variations in nuclear shape were not readily apparent in routine sections (fig. 6), these features became more evident in plasticembedded sections (fig. 1).
Nuclear Profile Invaginations Figure 4 displays graphically the percentages of" nuclear profiles of each morphotype with membrane invaginations of 0.4 ixm or more in depth for normal lymph node and large cell lymphoma specimens. Less than 25 per cent of nuclear profiles in all regions of normal lymph nodes hadP significant nuclear membrane indentations, while all specimens from large cell lymphomas had levels of indented profiles that 967
HUMAN PATHOLOGY
Volume '15, No. 10 (October 1984) 100
were considerably greater. In many lymphoma specimens, 40 to 80 per cent of the nuclear profiles for each of the three morphotypes had invaginations 0.4 p.m or more in depth. Nuclear profile invagination depth. Despite the considerable p r o p o r t i o n o f nuclear profiles with membrane indentations, it is evident from figure 5 that the majority o f large cell lymphomas had mean nuclear membrane invaginati0n depths of approximately the same size as those found in the nuclei of normal lymphocytes in germinal centers. In germinal follicle lymphocytes there was a progressive increase in the mean depth of invaginations (fig. 5), paral-
90
o
8
80 8
70
8
o
9 o
u 50
!
40 30
o
8 o
20 10
60 50 ~, 40
9
8
9
Oo o
N 3o
9 20
S
9
,.
~o
lJl
I
M1
GC1
I
I
I
I
I
LC1 GC2 LC2 GC3 NUCLEARMORPHOTYPE
2.5
9
2.0
0
0 o
u~ i.5 z
:
o
9 9
oo
L
L
LC3
o
1.0
t
9
OD
o0o o cD
9
o .....
10i
0.5 0
0
I
I
MI
GCI
I
I
[
f
LCI GC2 LC2 GC3 NUCLEARMORPHOTYPE
t
LC3
f
0
H
5.0
o
9
r~ 4.4 Z 4.2
Z O u 4.0
3,
. ~
s.
9
P'
P
9
I
I
MI
GCI
I
I
i" I
I
t
I
I
I
i
I
LC1 GC2 LC2 GC3 NUCLEARMORPHOTYPE
i
I
tO3
H
FIGURE 5 (bottom]. Mean nuclear invagination depth. Although neoplastic lymphocyte nuclear profiles for all morphotypes have a higher percentage of indentations (fig. 4), the mean depth of the invaginations is generally less than those of normal lymphocyte nuclear profiles.
3.6
3.4
I
GCI
FIGURE 4 (top]. Nuclear profile indentations. Each point represents the percentage of individual nuclear morphotype profiles with indentations of 0.4 p,m or more in depth in mantle and germinal center regions of the'normal lymph node and the 20 large cell lymphoma specimens. The relatively low level of nuclear membrane indentations for lymphocytes from normal lymph nodes contrasts with the generally higher levels In large cell lymphomas. The ranges of values in "normal, residual" (LCl] lymphocytes and neoplastic lymphocytes with morphotype 2 and 3 nuclei are similar.
4.8 4.6
I M1
LCI GC2 tC2 GC3 LC3 H NUCLEARMORPHOTYPE FIGURE 2 (top]. Mean nuclear area. Each point represents the mean value of a series of morphometric measurement of lymphocyte nuclei in selected regions of reactive lymph nodes and in large cell lymphornas. FIGURE 3 [bottom]. Mean nuclear contour index (NClI. The mean NCl of normal germinal center lymphocytes is generally in the range of 3.7 to 4.0, while the m e a n value for the majority of neoplastic l y m p h o c y t e nuclei of all morphotypes is considerably greater. The ranges of the mean values for all three nuclear morphotypes of large cell lymphomas [i.ev LCI, LC2, and LC3] are similar. M4, mantle zone, morphotype '1 lymphocyte nuclei; GCl, GC2, and GC3, germinal center, morphotype I, 2, and 3 lymphocyte nuclei, respectively;, LCl, LC2, and LC3, large cell lymphoma, morphotype 1, 2, and 3 nuclei, respectively, H, normal histiocyte nucleus. Solid circles indicate normal lymph nodes. Open circles represent lymphomas fixed in formalin and open diamonds those fixed in Karnovsky's solution.
leling the increase in mean nuclear area from morp h o t y p e l to m o r p h o t y p e 3 lymphocytes (fig. 2). However, the mean depths of the invaginations in m o r p h o t y p e 2 and 3 nuclei in the large cell lymphomas spanned similar ranges (fig. 5), again paralleling the similar ranges of mean nuclear area observed in neoplastic morphotype 2 and 3 nuclei in figure 2. DISCUSSION
In most surgical pathology practices, the mainstay of the classification 6f NHL remains the histologic patterns and cytologic characteristics evident in routine tissue sections. All current classification sys968
NUCLEAR ANALYSESIN NHL [Dardick et al.)
FIGURE 6. Large cell lymphoma, paraffin-embedded sections. Left,, diffuse type (case 1 in table 1]. Right, nodular variant [case 2 in table 1]. The difference in nuclear size between "residual" small lymphocytes and tumor cells is readily appreciated in both cases. It is more difficult to discern the difference in nuclear size of the neoplastic lymphocytes in the two large cell lymphomas [table I) with a x 25 objective. [Hematoxylin-eosin stain, x 350.]
tems are based on this principle, and no one system appears to have a distinct a d v a n t a g e over the others.38,41 With the advent of tile Lukes-Collins classifications,9-11 the immunologic aspects of N H L and normal lymphocyte transformation have been tile subject o f intensive investigation, but the unique basis for the morphologic aspects of this system has been largely untested. T h e hypothetic lymphocyte transformation sequence of small and large, cleaved and noncleaved cells that forms such an integral part of this classification system appears to have been designed to a c c o m m o d a t e cytologic observations in NHL. Until recently, no experimental observations of changes in nuclear size, shape, or condensed chromatin organization during the process of lymphocyte transformation in either B or T cells had been made. However, i n d u c t i o n o f h u m a n peripheral T-cell transformation in vitro 34 and mouse spleen B cells in vivo3~has revealed no specific phase of nuclear irregularity or clefting during the progressive increase in nuclear size from the small, unstimulated lymphocyte to the fully transformed state prior to DNA replication and cell division. Furthermore, as shown by mofphometric techniques, clefted nuclear profiles 21,42 account for less than 2 to 3 per cent of all stages of t r a n s f o r m i n g lymphocyte populations of germinal centers in reactive lymph nodes. 43 It is with this back969
ground that the morphometrically derived data of this report must be assessed. In a number of studies and trials subjective evaluation of N H L has proved unreliable in providing homogeneous diagnostic categories, 1-~ and this applies to large cell lymphomas as well. 4,12,15 Tile homogeneity of the nuclear characteristics in the various subtypes of malignant lymphoma has not been thoroughly assessed in terms of size, shape, and heterochromatin distribution, nor has the relation of'these important diagnostic features to normal lymph node lymphocyte populations. As is apparent from Figure 2, normal, lymphocyte nuclei (whether in mantle or germinal center regions) show considerable variation in mean nuclear area. In the case of germinal center lymphocytes, the ranges o f mean nuclear area in the TABLE 1. Mean Nuclear Area [I.Lm2 ___ SEM) of Lymphocyte Populations with Morphotype 1, 2, and 3 Nuclei in Two Large Cell Lymphomas [Case 1, Upper End of Size Range in fig. 2; Case 2, Lower End]
Nuclear Morphotype
Case 1
Case 2
LCI LC2 LC3
10.6 _ 0.3 35.0 _ 0.9 40.4 _ 0.6
11.5 - 0.3 21.6 - 0.4 21.1 _+ 0.3
HUMAN PATHOLOGY
Volume 15, No. 10 (October 1984)
three subtypes, based on nuclear morphology, appear to confirm the progressive alteration of small lymp h o c y t e nuclei to the fully t r a n s f o r m e d state following antigenic or mitogenic stimulation. 35 It is evident from morphometric assessment o f the group of large cell lymphomas that mean nuclear areas in this subgroup do not achieve the range of values for fully t r a n s f o r m e d lymphocytes and, in fact, parallel the mean nuclear areas of partially transformed lymphocytes. Except for differences in condensed chromatin distribution, morphotype 2 and 3 tumor cells in large cell lymphomas a p p e a r to represent a single neoplastic population. In addition to the similar ranges o f mean nuclear area, tile mean NCI, the mean invagination depth, and the percentage of nuclei with invaginations also span similar ranges in both nuclear classes. O f interest were the few large cell lymphomas with mean nuclear areas that fell within the range for normal small lymphocytes (fig. 2). Whitcomb et alJ 5 were also aware of a small number of diffuse histiocytic lymphomas in.which nuclei were relatively small but still appeared to be "transformed." Such results imply not necessarily errors in morphologic classification but rather tile inadequacy of the usual visual assessment of lymphomas. For example, tile difference in diameter between nuclei with areas of 25-and 40 ~m 2 is only 1.5 p.m. Such small differences in nuclear diameter are not readily appreciated at the microscopic magnifications usually used for assessing biopsies of lymphoid tissues (fig. 6). T h e spectrum of nuclear size in large cell lymphomas compares with a similar situation in lung tumors, for which there is also no clear-cut separation of small and large cell carcinomas. 44 Results of morphometric studies for comparison with the present series o f large cell lymphomas are scarce, but the values are generally within the same range.17,22. 24 In a retrospective study, such as the present investigation, it may be argued that differences in fixation and tissue processing may account for the wide range of nuclear size found. A number o f findings refute this argument. Van der Valk et al., 23,24using uniformly fixed and embedded tissue, found mean nuclear areas ranging from 27 to 57 ixm 2 in 16 B immunoblastic sarcomas and large noncleaved follicular center cell lymphomas. In addition, figure 2 reveals that the distributions of nuclear size for specimens fixed in formalin and Karnovsky's solution are similar. Since the nuclear size of reactive histiocytes has been the standard for segregating large and small cell lymphomas, it should be appreciated that the mean nuclear area of histiocytes as reported by van der Valk et al. 22 was 34.3 ___ 11.7 ixm~ and that a mean of 40.0 + 15.0 I-tm2 was found in the present study. In the present series many of the mean nuclear areas in large cell lymphomas were less than these values. An u n e x p e c t e d finding involved the nuclear characteristics of the small, unstimulated lymphocytes that formed a component: in each of the 20 large cell
lymphomas analyzed morphometrically (4 to 37 per cent o f the total population measured in each case; mean, 17 per cent). These lympocytes have tradiiionally been referred to as residual normal lymphocytes. However, morphometric assessment generally reveals that the small lymphocytes in large cell lymphomas are smaller than normal mantle zone and germinal center small lymphocytes and have irregularities of nuclear shape (NCI) that parallel those of the larger nucleated cells that are usually considered the neoplastic component. T h e exact quantity and topology of nonneoplastic B and T lymphocytes in N H L have not been definitively established. 45-49 T h e current findings suggest that at least some of these small lymphocytes are part o f the t u m o r cell cohort. T h e pathobiologic role and significance of this population remain to be clarified. It is possible tha t they represent part of a similar maturation mechanism used to explain h e t e r o g e n e o u s t u m o r cell populations in N H L o f the follicular center cell type.14A6,50 Despite lack o f understanding of the nature and role of small, unstimulated lymphocytes in large cell l y m p h o m a , their presence is the key to u n d e r standing the spectrum o f nuclear size that occurs in NHL. Both the size and chromatin distribution of the nuclei o f small lymphocytes contrast with the finely distributed, condensed chromatin in the nuclei of the larger tumor cell component of large cell lymphomas. However, it is primarily the size of the nuclei in the small lymphocytes that is responsible for creating the illusion of a "large" nucleated neoplastic cell in the large cell lymphomas in which the mean nuclear area is within the nuclear size range for normal small lymphocytes in mantle zones and germinal centers. Case 2 (table 1) illustrates this phenomenon: the mean nuclear area o f LC 1 (1 t.5 ~m 2) is considerably smaller than that of LC3 (21.2 I.tm2), but the area of the latter is within the range of nuclear area for GC1 lymphocytes shown in figure 2 (16.5 to 24.9 Ixm~). All current classifications o f N H L segregate the various subtypes according to the common denominators of nuclear size and condensed chromatin distribution. However, with the exception of tile recognition o f architectural patterns, current criteria used for categorizing the spectrum of N H L subtypes have been unreliable. 1-5 F u r t h e r m o r e , although it has increased knowledge of the developmental and functional aspects of lymphocytes, the immunologic approach to N H L classification has not enhanced diagnostic capabilities in the morphologic aspects of these disorders.Sl, 52 In terms of patient management, the current morphometric assessment of large cell lymphomas does not simply demonstrate inconsistencies in N H L classification criteria and reasons for int6robserver diagnostic discrepancies. T h e s e results also raise questions concerning the biologic basis for the current classification systems that are thought to reflect the morphologic stages in normal lymphocyte transformation. T h e acc~lracy of the latter hypothesis may be immaterial in view of the findings of the NonH o d g k i n ' s L y m p h o m a Pathologic Classification 970
NUCLEARANALYSESIN NHL [Dardick et al.]
Project. 38 F r o m the p r o g n o s t i c a n d therapeutic viewpoints, with this c l a s s i f i c a t i o n s y s t e m N H L appears to be divided into three b r o a d categories that reflect behavioral aspects, with an a p p a r e n t disregard for the systematic a r r a n g e m e n t o f subtypes in o t h e r classifications. I n light o f the p r e s e n t results indicating t h a t l a r g e cell l y m p h o m a s a r e n o t h o m o g e n e o u s with r e g a r d to n u c l e a r size a n d gross c h r o matin organization, it is possible that nuclear size is not as i m p o r t a n t a prognostic indicator or diagnostic criterion as stressed previously. Clearly, n o t all large cell l y m p h o m a s are truly c o m p o s e d o f neoplastic lymphocytcs with "large" nuclei. More i m p o r t a n t is the visual effect created by the distribution o f c o n d e n s e d c h r o m a t i n . It is obvious that accurate evaluation o f nuclear size in histologic sections by routine observation is difficult. U n f o r t u n a t e l y , visual assessment o f c o n d e n s e d c h r o m a t i n a m o u n t s a n d d i s t r i b u t i o n is also unreliable. 53-56 C l i n i c o p a t h o l o g i c studies o f d i f f u s e histiocytic l y m p h o m a have not s u c c e e d e d in establishing significantly d i f f e r e n t o r c o n s i s t e n t p a t t e r n s o f survival data f o r the various m o r p h o l o g i c subtypes. 4,6,1~,1~,15 However, W a r n k e et al. 4 indicated that it m a y be possible to segregate large cell l y m p h o m a s into two b r o a d prognostic categories. I n s u m m a r i z i n g the literature and their o w n results, these a u t h o r s 4 questioned the" u s e f u l n e s s o f f r a g m e n t i n g d i f f u s e histiocytic lymp h o m a into n u m e r o u s m o r p h o l o g i c subtypes. Because o f the difficulties e n c o u n t e r e d even in distinguishing "small" f r o m "large" nucleated l y m p h o m a s in the c u r r e n t study, this seems to be a reasonable assessment. Perhaps i m p r o v e m e n t s in m o r p h o l o g i c criteria a n d modifications o f seemingly i n a p p r o p r i a t e t e r m i n o l o g y in existing N H L classifications can result f r o m a systematic a n d objective analysis o f neoplastic l y m p h o c y t e s in the various subcategories. Certainly, the biochemical a n d cell biologic aspects o f nuclear organization in n o r m a l a n d l y m p h o m a t o u s l y m p h o cytes must be investigated m o r e t h o r o u g h l y and understood. Results o f clinical trials will be m e a n i n g f u l only w h e n the pathobiologic processes resulting in the diverse s p e c t r u m o f t u m o r o u s lymphocytes are u n d e r s t o o d a n d m e a n i n g f u l a n d r e p r o d u c i b l e cytologic criteria are established.
Acknowledgment. The authors are indebted to the Canadian Pathologists, who contributed case material to tiffs study. REFERENCES 1. Ezdinli EZ, Costell0 W, Wasser LP, et al: Eastern Cooperative Oncology Group experience with the Rappaport classification of non-Hodgkin's lymphomas. Cancer 43:544, 1979 92. Jones SE, Butler JJ, Byrne GE, et al: Histopathologic review of lymphoma cases from the Southwest Oncology Group. Cancer 39:1071, 1977 3. Kim H, Zelman RJ, Fox MA, et al: Pathology panel for lymphoma clinical studies: a comprehensive analysis of cases accumulated since its inception. J Nail Cancer Inst 68:43, 1982 4. Warnke RA, Strauchen JA, Burke ]S, et ah Morphologic types of diffuse large-cell lymphoma. Cancer 50:690, 1982
971
5. Woodruff RD: Reviewing histologic diagnosis in lymphoma. Arch Pathol Lab Med 105:573, 1981 6. Armitage JO, Dick FR, Platz CE, et ah Clinical usefulness and: reproducibility of histologic subclassification of advanced diffuse histiocytic lymphoma. AmJ Med 67:929, 1979 7. Azar HA, Jaffe ES, Berard CW, et al: Diffuse large cell lymphomas (reticulum cell sarcomas, histiocytic lymphomas). Correlation of morphologic features with functional markers. Cancer 46:1428, 1980 8. Berard CW, Jaffe ES, Braylan RC, et al: Immunologic aspects and pathology of tile malignant lymphomas. Cancer 42:911, 1978 9. Lukes RJ, Collins RD: Immunologic characterization of human malignant lymphomas. Cancer 34:1488, 1974 I0. Lukes RJ, Collins RD: New approaches to the classification of the lymphomata. BrJ Cancer 31 (suppl II): 1, 1975 11. LukesRJ, ParkerJW, Taylor CR, et al: Immunologic approach to non-Hodgkin's lymphomas and related leukemias..Analysis of the results of muhiparameter studies of 425 cases. Semin Hematol 15:322, 1978 12. Newcomer LN, Nerenberg MI, Cadman EC, et al: The usefulness of the Lukes-Collins classification in identifying subsets of diffuse histiocytic lymphoma responsive to chemotherapy. Cancer 50:439, 1982 13. Strauchen JA, Young RC, DeVita VT, et ah Clinical relevance of the histopathological subclassification of diffuse "histiocytic" lymphoma. N EnglJ Med 299:1382, 1978 14. Taylor CR: Results of multiparameter studies of B-cell lymphomas. AmJ Clin Pathol 72:687, 1979 15. Whitcomb CC, Cousar JB, Flint A, et al: Subcategories of histiocytic lymphoma: associations with survival and reproducibility of classification. The Southeastern Cancer Study Group experience. Cancer 48:2464, 1981 16. Maurer R, Taylor CR, Parker JW, et al: Immunoblastic sarcoma: morphologic criteria and the distinction of B and T cell types. Oncology 39:42, 1982 17. Abbott CR, Blewitt RW, Bird CC: Quantitative analysis ofnonHodgkin's lymphoma. J Clin Pathol 35:135, 1982 18. Crocker J, Curran RC: A study of nuclear diameters in lymph node imprints using the Zeiss Microvideomat.J Clin Pathol 32:670, 1979 19. Crocker J, Jones EL, Curran RC: Study of nuclear diameters in non-Hodgkin's lympho.mas.J Clin Pathol 35:954, 1982 20. Crocker J, Jones EL, Curran RC: A comparative study of nu9clear form factor, area and diameter in non-Hodgkin's lymphomas and reactive lymph nodes. J Clin Pathol 36:298, 1983 21. EimotoT, Mitsui T, Kikuchi M: Ultrastructure of adult T-cell leukemia/lymphoma. Virchows Arch [Cell Pathol] 38:189, 1981 22. van der Valk P, te Velde J, Jansen J, et al: Malignant lymphoma of true histiocytic origin: a morphological, ultrastructural, cytochemical and clinical study of 10 cases. Virchows Arch [Pathol Anat] 391:249, 1981 23. van der Valk P, Hermans J, Brand R, et al: Morphometric characterization of diffuse large-cell (histiocytic) lymphomas. Am J Pathol 107:327, 1982 24. van der Valk P, Mosch A, Kurver PJ, et ah Morphometric characterisation of 52 B cell non-Hodgkin's lymphomas. J Clin Pathol 36:289, 1983 25. Caorsi I, Figueroa CD, Rodriguez EM: Morphologic and morphometric study of the two main cell lineages involved in mycosis fungoides: the lymphoid cells and Langerhans cells. Uhrastruct Pathol 3:119, 1982 26. Litovitz TL, Lutzner MA: Quantitative measurements of broad lymphocytes from patients with chronic lymphocytic leukemia and the S6zary syndrome. J Natl Cancer Inst 53:75, 1974 27. van der Loo EM, Cornelisse CJ, van Vloten WA, et al: Diagnostic morphometry of isolated lymph node cells from patients with mycosis fungoide s-and S6zary's syndrome. Virchows Arch [Cell Pathol] 33:107, 1980 28. van der Loo EM, van Vloten WA, Cornelisse cJ, et al: The
HUMAN PATHOLOGY
29. 30. 31. 32.
33. 34.
35.
36. 37. 38.
39. 40. 41. 42.
Volume 15, No. 10 (October 1984) 43. Dardick I: Morphometry of normal human lymphoid tissues. Nuclear invaginations and clefts. Arch Pathol Lab Med 108:197, 1984 44. Vollmer RT: Tile effect of cell size on the pathologic diagnosis of small and large cell carcinomas of the lung. Cancer 50:1380, 1982 45. Jaffe ES, Braylan RC, Nanba K, et ah Functional markers: a new perspective on malignant lymphomas. Cancer Treat Rep 61:953, 1977 46. Warnke R, Levy R: Detection of T and B cell antigens with hybridoma monoclonal antibodies: a biotin-avidin-horseradish peroxidase method. J Histochem Cytocllem 28:771, 1980 47. van Heerde P, Fehkamp CA, Fehkamp-Vroom TM, et al: Non-Hodgkin's lymphoma. Immunohistochemicaland electron microscopical findings in relation to light microscopy. A study of 74 cases. Cancer 46:2210, 1980 48. Harris NL, Data RE: The distribution of neoplastic and normal B-lymphoid cells in nodular lymphomas: use of an immunoperoxidase technique on frozen sections. Hv.xl PATHOL 13:610, 1982 49. Dvoretskey P, Wood GS, Lewvy R, et al: T-lymphocyte subsets in follicular lymphomas compared with those in non-neoplastic lymph nodes and tonsils. Hum Pathol 13:618, 1982 50. van den Tweel JG, Lukes RJ, Taylor CR: Pathophysiology of lymphocyte transformation. A study of so-called coutposite lyntphomas. Am J Clin Pathol 71:509, 1979 51. Bain GO: Non-Hodgkin's l)unphomas. Analysis of 92 cases using the "International" classification. Arch Pathol Lab Med 107:64, 1983 52. Jaffe ES, Strauchen JA, Berard CW: Predictability of immunologic phenotype by morphological criteria in diffuse aggressive non-Hodgkin's lymphoma. AmJ Clin Pathol 77:46, 1982 53. Dardick I, Setterfield G: Vohnne of condensed chromatin in developing primitive-lineerythrocytes of chick. Exp Cell Res 100:159, 1976 54. Dardick I, Sheinin R, Setterfield G: Mutant mouse L-cells: a model for megaloblastic anaemia. Br J Haematol 39:483, 1978 55. Setterfield G, Sheinin R, Dardick I, et al: Structure of interphase nuclei in relation to the cell cycle. J Cell Biol 77:246, 1978 56. Sheinin R, Setterfield G, Dardick I, et al: Relationship between chromatin structure and replication in ts mouse L-cells. Can J Biochem 58:1359, 1980
relevance of morphometry in the differential diagnosis of cutaneous T cell lymphomas. BrJ Dermatol 104:257, 1981 Lutzner M, Edelson R, Shein P, et al: Cutaneous T-cell lymphomas: the Sdzary syndrome, mycosis fungoides and related disorders. Ann Intern Med 83:534, 1975 McNutt NS, Heilbron DC, Crain WR: Mycosis fungoides. Diagnostic criteria based on quantitative electron microscopy. Lab Invest 44:466, 1981 McNutt NS, Crain WR: Quantitative electron microscopic comparison of'lymphocyte nuclear contours in mycosis fungoides and inbenign infiltrates in skin. Cancer 47:698, 1981 Meijer CJLM, van der Loo EM, van Vloten WA, et al: Early diagnosis of mycosis fungoides and Sdzary's syndrome by morphometric analysis of lymphoid cells in the skin. Cancer 45:2864, 1980 Scheffer E, Meijer CJLM, van Vloten WA: Dermatopathic lymphadenopathy and lymph node involvement in mycosis fungoides. Cancer 45:137, 1980 Dardick I, Setterfield G, Hall R, et al: Nuclear aherations during l)mphocyte transformation. Relationship to the heterogeneous morphologic presentations of non-Hodgkin's lyntphomas. Am J Pathol 103:10, 1981 Dardick I, Sinnott NM, Hall R, et al: Nuclear morphology and morphometry of B lymphocyte transformation. Implications for follicular center cell lymphomas. Am J Pathol 111:35, 1983 Levine AM, Tayl~or,CR, Schneider DR, et al: Immunoblastic sarcoma of T-cell versus B-cell origin: I. clinical features. Blood 58:52, 1981 Lichtenstein A, Levine AM, Lukes RJ, et al: Inmtunoblastic sarcoma. A clinical description. Cancer 43:343, 1979 Report of the Writing Committee: National Cancer Institute sponsored study of classifications of non-Hodgkin's" lymphomas: summary and description of a working formulation for clinical use. Cancer 49:2112, 1982 Troyer H, Babich E: A hematoxylin and eosin-like stain for glycol methacrylate embedded tissue sections. Stain Technol 56:39, 1981 Schrek R: Ultrastructure of blood lymphocytes from chronic lymphocytic and lymphosarcoma cell leukemia. J Natl Cancer Inst 48:51, 1972 Nathwani BN: A critical analysis of the classifications of nonHodgkin's lymphomas. Cancer 44:347, 1979 Said JW, Hargreaves HK, Pinkus GS: Non-Hodgkin's lymphomas: an ultrastructural stud)' correlating morphology with immunologic cell type. Cancer 44:504, 1979
972
Both morphologically and immunologically, non.Hodgkin's lymphoma (NHL) of the large cell type has been shown to be a heterogeneous category. However, the homogeneity of the nuclear p a r a m e t e r s , particularly size and condensed chromatin organization, used to classify this subtype of NHL has not been investigated. In fact, objective morphologic techniques have not been systematically applied to verify the segregation of NttL on the basis of nuclear parameters, a concept common to all current classification systems. In this study morphometric image analysis was used to compare the nuclei in 20 specimens from NHLs of the large cell type with.those in mantle zone and germinal center lymphocytes from lymph nodes with reactive hyperplasia. R e s u l t s of the assessment of mean nuclear area in large cell lymphomas revealed that this class is also heterogeneous, with s o m e of the specimens having a nuclear size in the u p p e r range of that for normal small lymphocytes. In addition, in only a few of these specimens was the mean nuclear area within the range of that for fully transformed germinal center lymphocytes. The majority of large cell lymphomas have a nuclear size more characteristic of partially transformed lymphocytes in germinal centers. In addition to indicating inconsistencies in the current diagnostic criteria used in NHL classifications, the results indicate r e a s o n s for interobserver variations in clinicopathologic trials; the validity of nuclear size as a prognostic indicator and the biologic basis for classifying NHL as a reflection of normal lymphocyte transformation are also questioned. In terms of patient management, the classifications of NHL currently used require objective reappraisal. HUM PATHOL 15:965--972, 1984.
Recurring problems in clinical trials and pathologic studies o f n o n - H o d g k i n ' s l y m p h o m a (NHL) have been the lack of reproducibility and the subjective bias inherent in such studies. 1-5 Ahhough immunologic, morphologic, and prognostic studies have confirmed the heterogeneity of the large cell class of NHL, 6-16 difficulties in standardizing the specific morphologic criteria and definitions for each subcategory of large cell lymphoma persist. 4,6 One of the major reasons for the inconsistencies in morphologic diagnoses in NHL has been the reliance on subjective evaluation o f the cytologic features of neoplastic lymphocytes. Despite the important role of nuclear morphology in current classification systems, this organelle has been evaluated objectively only recently in some categories of NHL. 17-24 A more
MATERIALSAND METHODS
Material for this study was selected from cases that had been accessioned at the Canadian T u m o u r Reference CEntre, Ottawa, Canada, with diagnoses of diffuse or nodular histiocytic lymphoma and in which wet tissue or plastic resin-embedded tissue was available. Cases were examined independently by three of the authors (I.D., W.T.E.Mc. M.A.-J.), and in each a
Received October 7, 1983, from the Canadian Tumour Reference Centre, Ottawa, Ontario, Canada. Accepted for publication N o v e m b e r 17, 1983. Supported by a grant from the National Cancer Institute of Canada. Address correspondence and reprint requests to Dr. Dardick: Canadian T u m o u r Referenc.e Centre, Clinical Studies Unit Building, 60 Ruskin Avenue, Ottawa, Ontario, Canada KIY 4M9.
965
HUMAN PATHOLOGY
Volume 15, No. 10 [October 1984)
consensus diagnosis was established. By criteria established in the literature, 9-11,36-38 neoplasms composed primarily of large or medium-sized lymphoreticular cells that had the characteristics and cytologic details of large noncleaved cell lymphoma or immunoblastic sarcoma were segregated for morphometric analysis. Routinely p r e p a r e d histologic sections showed these tumors to be composed basically of neoplastic lymphocytes with generally smoothly contoured vesicular nuclei; one to three acidophilic nucleoli; and multiple, small, diffusely a r r a n g e d clumps of condensed chromatin. Tumors that had any significant component (>20 per cent) of small cleaved cells, a n d that might thus be classified as mixed, were excluded. Of the 20 tumors measured, 16 were diffuse large noncleaved cell lymphomas, two were nodular'large noncleaved cell lymphomas, and two were immunoblastic sarcomas. For comparative purposes, six lymph node biopsy specimens with a diagnosis of reactive hyperplasia and in which wet tissue was available were also selected for morphometric analysis. Histiocyte nuclear parameters were measured in prominent aggregates of this cell type in a lymph node involved.with atypical lymphoid hyperplasia. For both the lymphomas and the lymph nodes with reactive hyperplasia, the majority of tissues had been fixed in 10 per cent buffered formalin. One large cell lymphoma had been fixed in B-5, while seven specimens had been fixed initially in Karnovsky's solution. Wet tissue was processed for glycol . methacrylate resin e m b e d d i n g (Sorvall e m b e d d i n g medium, D u p o n t Instruments, Wilmington, Delaware), and sections were cut 1 p~m thick with a Sorvall JB4 microtome with glass knives and stained with celestine blue 39 or Gill's hematoxylin (Fisher Scientific, Pittsburgh, Pennsylvania) and eosin. Tissues fixed in Karnovsky's solution had been embedded previously in E p o n - A r a l d i t e resin; with a LKB IV ultramicrotome, 1-1~m sections were cut and stained with tolu{dine blue. For morphometric analysis, random but representative regions of the tumors and representative areas o f the mantle zoiae and germinal centers o f reactive lymph nodes were p h o t o g r a p h e d in nonoverlapping frames on 35-mm film with an oil immersion objective (100 x ). Each frame of the negative film was projected onto a digitizer pad (Summagraphics Corporation, Fairfield, Connecticut) at a final magnification of 2500 x . In each frame all welloutlined lymphocyte nuclear profiles were traced, and the data were collected and analyzed with a MSA II program (Atlantis Scientific Systems Group, Ottawa, Canada) and an Apple II Plus microcomputer (Apple Computer, Cupertino, California). The digitized information was used to calculate area, perimeter, volume, and contour index of each nuclear profile. For profiles having well-defined angulated or narrow identations, the depth of this identation was measured from the margins of the nuclear membrane on either side of the invagination to its base.
Contour index 4~ is a size-independent measurement of the shape of a profile in which increasing irregularity or eccentricity o f a profile results in ihigher values (a circle has a contour index of 3.54). T h e chromatin organization of normal and neoplastic nuclei was assessed by descriptive criteria reported previously. 34,35 Essentially, lymphocyte nuclei are subdivided into three types (fig. 1, left): morphotype 1 involves the prominent and compact masses of c o n d e n s e d c h r o m a t i n seen in u n s t i m u l a t e d small lymphocytes; morphotype 3 demonstrates the markedly disaggregated condensed chromatin and prominent nucleolus characteristic of the large, fully transformed lymphocyte; and morphotype 2 is characterized by nuclei with numerous, small to moderately sized clumps of condensed chromatin and thus represents the intermediate degree of condensed chromatin disaggregation a p p a r e n t in partially transformed lymphocytes.34,35 RESULTS
Each data point in figures 2 to 5 represents the mean value for a particular nuclear parameter measured morphometrically in each sample of lymphoid tissue. In the six normal but reactive lymph node biopsy Specimens, a total of 11,618 lymphocyte nuclei were measured: 4,160 m o r p h o t y p e 1 mantle zone (M1) l y m p h o c y t e nuclei (mean per sample, 693; range, 426 to 1,051); 1,800 morphotype 1 germinal center (GC1) lymphocyte nuclei (mean per sample, 300; range, 194 to 434); 4,381 morphotype 2 germinal center (GC2) nuclei (mean per sample, 730; range, 501 to 1,106); and 1,277 morphotype 3 germinal center (GC3) nuclei (mean per sample, 212; range', 107 to 519). Nuclear profiles traced in the 20 samples from large cell lymphomas totalled 41,020. Of these, 7,122 were categorized as m o r p h o t y p e 1 (LC1) nuclei (mean per sample, 356; range, 41 to 732); 12,544 as morphotype 2 (LC2) (mean per sample, 672; range, 335 to 997); and 21,354 as m o r p h o t y p e 3 (LC3) (mean per sample, 1,068; range, 521 to 1,708). Nuclear Profile A r e a
For comparative purposes, figure 2 illustrates tile ranges o f mean nuclear profile areas for normal, small-unstimulated lymphocytes (morphotype 1) in mantle zones and germinal centers, and for partially transformed (morphotype 2) and fully transformed (morphotype 3) lymphocytes of germinal centers. For typical neoplastic lymphocytes in large cell lymphomas (LC2 and LC3), mean nuclear size ranged from 12 to 50 p.m 2. Even morphotype 3 lymphocytes (LC3), with their open, vesicular nuclei, had a nuclear size range of 18 to 50 ~m~ (fig. 2). Unexpectedly, the mean nuclear profile areas for small morphotype 1 lymphocytes in the large cell lymphomas did not fall within tile range of morphotype 1 nuclei for normal tissues, but were generally smaller. Similarly, the majority of mean nuclear areas for morphotype 3 tumor 966
NUCLEAR ANALYSESIN NHL CDardick et al.l
k~,.';,.-r
+" 2 : - - ", ' + ~ r +4:'~+;
)
"
.:'
"~-',"
:
~,~-.~
,
-
;+
-
,
.
.,-~ "---"
"
~ .
",-'')2 "-,+
~:~:~
,
,~.:;4
,~" ,r . . . .
+
+.
9
l+
",r~)+,
,:.
s;~
"~
~-./.
,+
,.;
" .:
*.% 9
"
'
"-':
.
;"
C.'":.':~-._)'.~ .. -,,.
.
. ~ . , . ~ ,
:~.
.... ..
-.,
.-~
"
O" _a,_
-
.J
~:-~.
. -,~.
+..,,
.+,
'~t.'i..:
~
+,, ,
;': "
-~'~ll~'..-'_-Y~".v,
"
,LJ",
r
"~'.
". '
:~
.
-+-\7
~
C-,.
~.,-~
,.r ;._..
, I~ + . , + 3 ' 9 .. . ,.+.7:> ~_:1--~ +-+-* .': 9 "-+'.+-~+_~,. ".~ ---:++" . " . .' . . " "'r, .... ~" : ' " " * " "'~" -+','"+"~.."
'
...
~ - - ..
~.+,';+'
.
iai~
.
."~'~..'~ ",%~+'+~5.."
'-:
g.,j
,~---. ~ i , ~ . ' - ~ , , I, ~ 3 " " ... ,. L.++'U;" [ +i. _2 *'~/~-~ID . 9 ", + " . . ~. ;. , .
.
~ ' ~
+
. .,,,
9
- +" " .
. . . .
'~.
~
.""~ "+<'~ ~
!1~
.
,.,,.'i~It
?r,-L.,...-+.),
.._:, . .
"~'"" +,',,.+, .'~k.+'r"
-,
. [~..'. 9" ' ;
t~ 2. ,)
-'--~'~ ..
,-v .-.
--':
TM
e_~
i,,,,
f.,.. : p.+j
-~ .+
9
. . . . .
).
- . J
"
, ,'.
~r~.," 9 - ,
, ~:.'=-, (-~" . . . .
f~.,
(_~
~
..'v.
~# '" ~ - ' -
-,---.~-~
"
e~..
,"
"t-+---
-++'++., "+ b
v.--~:
"-,,
9 "+,
; , /-.~
"
/~;-".-~
~;
~-
~.-~--,,I~. -+ ~ ' ~
_-5--.~
......
I
.
,+.,"
' '
9 -
+..=',
, . :-,.. --:.-'~: , - ' . J_'f,Z T;~',-.': , . , + , ~ 2 < 3 . :2" : - . ' - " , , . "_ c-~,w.._~+.~_.: +.i; , - - z ~ ~'~" _'~; FIGURE 1. Large cell lymphoma, Epon-embedded sections Left, diffuse lymphoma from case 1. Right, nodular lymphoma from case 2. Examination with a x 40 objective of l-i~m sections clarifies the mean nuclear area differences recorded for the two cases presented in table I. Numerals designate typical examples of the three nuclear morphotypes. (Toluidine blue stain, x 550.)
cells in large cell lymphomas were less than the lowest mean nuclear area values for morphotype 3 lymphocytes in normal germinal centers (fig. 2). In fact, a n u m b e r of the large cell lymphomas had mean nuclear areas within the range for morphotype 1 lymphocytes of the mantle zone and germinal center. Although varying in condensed chromatin distribution, m o r p h o t y p e 2 and 3 nuclei in the large cell lymphomas had similar ranges of mean profile area. It is evident that mean nuclear area values for formalinand Karnovsky's solution-fixed t u m o r tissues were similar (fig. 2). Figures 1 (left) and 6 (le/t) represent regions from one of the large cell lymphomas in which the areas of neoplastic lymphocyte nuclei were in the upper range, while figures 10ight) and 6 (right) demonstrate the lower range of mean nuclear area (fig. 2 and table 1). Figure 6 shows paraffin-embedded tissue photographed with a x 2 5 objective, and figure 1 shows E p o n - A r a l d i t e plastic-embedded sections photographed with a x 40 objective. In the former it was difficult to appreciate differences between the two examples in nuclear area; the differences were more readily apparent in the plastic-embedded tissues. In both large cell lymphomas (fig. 6, left and right), it was easier to detect differences in nuclear size between the lymphocytes with small, darkly staining nuclei (LC1) and the obvious neoplastic nuclei (LC2 and LC3). T h e similarity of the mean nuclear areas of LCl-type cells in these two'large cell lymphomas and
the differences between tile LC1 population of lymo phocytes and the LC2- and LC3-type cells in case 2 are evident in table 1.
Nuclear Contour Index [NCI) A narrow range of NCI, and only minimal deviation from a circular outline, characterized the nuclei of normal lymphocytes, whether in the mantle zone or germinal centers (fig. 3). All three nuclear subtypes in large cell lymphomas also had similar ranges o f m e a n NCI. However, unlike those in n o r m a l lymphocytes, n u c l e a r profiles in the lymphomas were generally considerably more irregular, as reflected by the significantly higher mean NCI of the majority of specimens (fig. 3). Ahhough the degrees of-irregularity and the variations in nuclear shape were not readily apparent in routine sections (fig. 6), these features became more evident in plasticembedded sections (fig. 1).
Nuclear Profile Invaginations Figure 4 displays graphically the percentages of" nuclear profiles of each morphotype with membrane invaginations of 0.4 ixm or more in depth for normal lymph node and large cell lymphoma specimens. Less than 25 per cent of nuclear profiles in all regions of normal lymph nodes hadP significant nuclear membrane indentations, while all specimens from large cell lymphomas had levels of indented profiles that 967
HUMAN PATHOLOGY
Volume '15, No. 10 (October 1984) 100
were considerably greater. In many lymphoma specimens, 40 to 80 per cent of the nuclear profiles for each of the three morphotypes had invaginations 0.4 p.m or more in depth. Nuclear profile invagination depth. Despite the considerable p r o p o r t i o n o f nuclear profiles with membrane indentations, it is evident from figure 5 that the majority o f large cell lymphomas had mean nuclear membrane invaginati0n depths of approximately the same size as those found in the nuclei of normal lymphocytes in germinal centers. In germinal follicle lymphocytes there was a progressive increase in the mean depth of invaginations (fig. 5), paral-
90
o
8
80 8
70
8
o
9 o
u 50
!
40 30
o
8 o
20 10
60 50 ~, 40
9
8
9
Oo o
N 3o
9 20
S
9
,.
~o
lJl
I
M1
GC1
I
I
I
I
I
LC1 GC2 LC2 GC3 NUCLEARMORPHOTYPE
2.5
9
2.0
0
0 o
u~ i.5 z
:
o
9 9
oo
L
L
LC3
o
1.0
t
9
OD
o0o o cD
9
o .....
10i
0.5 0
0
I
I
MI
GCI
I
I
[
f
LCI GC2 LC2 GC3 NUCLEARMORPHOTYPE
t
LC3
f
0
H
5.0
o
9
r~ 4.4 Z 4.2
Z O u 4.0
3,
. ~
s.
9
P'
P
9
I
I
MI
GCI
I
I
i" I
I
t
I
I
I
i
I
LC1 GC2 LC2 GC3 NUCLEARMORPHOTYPE
i
I
tO3
H
FIGURE 5 (bottom]. Mean nuclear invagination depth. Although neoplastic lymphocyte nuclear profiles for all morphotypes have a higher percentage of indentations (fig. 4), the mean depth of the invaginations is generally less than those of normal lymphocyte nuclear profiles.
3.6
3.4
I
GCI
FIGURE 4 (top]. Nuclear profile indentations. Each point represents the percentage of individual nuclear morphotype profiles with indentations of 0.4 p,m or more in depth in mantle and germinal center regions of the'normal lymph node and the 20 large cell lymphoma specimens. The relatively low level of nuclear membrane indentations for lymphocytes from normal lymph nodes contrasts with the generally higher levels In large cell lymphomas. The ranges of values in "normal, residual" (LCl] lymphocytes and neoplastic lymphocytes with morphotype 2 and 3 nuclei are similar.
4.8 4.6
I M1
LCI GC2 tC2 GC3 LC3 H NUCLEARMORPHOTYPE FIGURE 2 (top]. Mean nuclear area. Each point represents the mean value of a series of morphometric measurement of lymphocyte nuclei in selected regions of reactive lymph nodes and in large cell lymphornas. FIGURE 3 [bottom]. Mean nuclear contour index (NClI. The mean NCl of normal germinal center lymphocytes is generally in the range of 3.7 to 4.0, while the m e a n value for the majority of neoplastic l y m p h o c y t e nuclei of all morphotypes is considerably greater. The ranges of the mean values for all three nuclear morphotypes of large cell lymphomas [i.ev LCI, LC2, and LC3] are similar. M4, mantle zone, morphotype '1 lymphocyte nuclei; GCl, GC2, and GC3, germinal center, morphotype I, 2, and 3 lymphocyte nuclei, respectively;, LCl, LC2, and LC3, large cell lymphoma, morphotype 1, 2, and 3 nuclei, respectively, H, normal histiocyte nucleus. Solid circles indicate normal lymph nodes. Open circles represent lymphomas fixed in formalin and open diamonds those fixed in Karnovsky's solution.
leling the increase in mean nuclear area from morp h o t y p e l to m o r p h o t y p e 3 lymphocytes (fig. 2). However, the mean depths of the invaginations in m o r p h o t y p e 2 and 3 nuclei in the large cell lymphomas spanned similar ranges (fig. 5), again paralleling the similar ranges of mean nuclear area observed in neoplastic morphotype 2 and 3 nuclei in figure 2. DISCUSSION
In most surgical pathology practices, the mainstay of the classification 6f NHL remains the histologic patterns and cytologic characteristics evident in routine tissue sections. All current classification sys968
NUCLEAR ANALYSESIN NHL [Dardick et al.)
FIGURE 6. Large cell lymphoma, paraffin-embedded sections. Left,, diffuse type (case 1 in table 1]. Right, nodular variant [case 2 in table 1]. The difference in nuclear size between "residual" small lymphocytes and tumor cells is readily appreciated in both cases. It is more difficult to discern the difference in nuclear size of the neoplastic lymphocytes in the two large cell lymphomas [table I) with a x 25 objective. [Hematoxylin-eosin stain, x 350.]
tems are based on this principle, and no one system appears to have a distinct a d v a n t a g e over the others.38,41 With the advent of tile Lukes-Collins classifications,9-11 the immunologic aspects of N H L and normal lymphocyte transformation have been tile subject o f intensive investigation, but the unique basis for the morphologic aspects of this system has been largely untested. T h e hypothetic lymphocyte transformation sequence of small and large, cleaved and noncleaved cells that forms such an integral part of this classification system appears to have been designed to a c c o m m o d a t e cytologic observations in NHL. Until recently, no experimental observations of changes in nuclear size, shape, or condensed chromatin organization during the process of lymphocyte transformation in either B or T cells had been made. However, i n d u c t i o n o f h u m a n peripheral T-cell transformation in vitro 34 and mouse spleen B cells in vivo3~has revealed no specific phase of nuclear irregularity or clefting during the progressive increase in nuclear size from the small, unstimulated lymphocyte to the fully transformed state prior to DNA replication and cell division. Furthermore, as shown by mofphometric techniques, clefted nuclear profiles 21,42 account for less than 2 to 3 per cent of all stages of t r a n s f o r m i n g lymphocyte populations of germinal centers in reactive lymph nodes. 43 It is with this back969
ground that the morphometrically derived data of this report must be assessed. In a number of studies and trials subjective evaluation of N H L has proved unreliable in providing homogeneous diagnostic categories, 1-~ and this applies to large cell lymphomas as well. 4,12,15 Tile homogeneity of the nuclear characteristics in the various subtypes of malignant lymphoma has not been thoroughly assessed in terms of size, shape, and heterochromatin distribution, nor has the relation of'these important diagnostic features to normal lymph node lymphocyte populations. As is apparent from Figure 2, normal, lymphocyte nuclei (whether in mantle or germinal center regions) show considerable variation in mean nuclear area. In the case of germinal center lymphocytes, the ranges o f mean nuclear area in the TABLE 1. Mean Nuclear Area [I.Lm2 ___ SEM) of Lymphocyte Populations with Morphotype 1, 2, and 3 Nuclei in Two Large Cell Lymphomas [Case 1, Upper End of Size Range in fig. 2; Case 2, Lower End]
Nuclear Morphotype
Case 1
Case 2
LCI LC2 LC3
10.6 _ 0.3 35.0 _ 0.9 40.4 _ 0.6
11.5 - 0.3 21.6 - 0.4 21.1 _+ 0.3
HUMAN PATHOLOGY
Volume 15, No. 10 (October 1984)
three subtypes, based on nuclear morphology, appear to confirm the progressive alteration of small lymp h o c y t e nuclei to the fully t r a n s f o r m e d state following antigenic or mitogenic stimulation. 35 It is evident from morphometric assessment o f the group of large cell lymphomas that mean nuclear areas in this subgroup do not achieve the range of values for fully t r a n s f o r m e d lymphocytes and, in fact, parallel the mean nuclear areas of partially transformed lymphocytes. Except for differences in condensed chromatin distribution, morphotype 2 and 3 tumor cells in large cell lymphomas a p p e a r to represent a single neoplastic population. In addition to the similar ranges o f mean nuclear area, tile mean NCI, the mean invagination depth, and the percentage of nuclei with invaginations also span similar ranges in both nuclear classes. O f interest were the few large cell lymphomas with mean nuclear areas that fell within the range for normal small lymphocytes (fig. 2). Whitcomb et alJ 5 were also aware of a small number of diffuse histiocytic lymphomas in.which nuclei were relatively small but still appeared to be "transformed." Such results imply not necessarily errors in morphologic classification but rather tile inadequacy of the usual visual assessment of lymphomas. For example, tile difference in diameter between nuclei with areas of 25-and 40 ~m 2 is only 1.5 p.m. Such small differences in nuclear diameter are not readily appreciated at the microscopic magnifications usually used for assessing biopsies of lymphoid tissues (fig. 6). T h e spectrum of nuclear size in large cell lymphomas compares with a similar situation in lung tumors, for which there is also no clear-cut separation of small and large cell carcinomas. 44 Results of morphometric studies for comparison with the present series o f large cell lymphomas are scarce, but the values are generally within the same range.17,22. 24 In a retrospective study, such as the present investigation, it may be argued that differences in fixation and tissue processing may account for the wide range of nuclear size found. A number o f findings refute this argument. Van der Valk et al., 23,24using uniformly fixed and embedded tissue, found mean nuclear areas ranging from 27 to 57 ixm 2 in 16 B immunoblastic sarcomas and large noncleaved follicular center cell lymphomas. In addition, figure 2 reveals that the distributions of nuclear size for specimens fixed in formalin and Karnovsky's solution are similar. Since the nuclear size of reactive histiocytes has been the standard for segregating large and small cell lymphomas, it should be appreciated that the mean nuclear area of histiocytes as reported by van der Valk et al. 22 was 34.3 ___ 11.7 ixm~ and that a mean of 40.0 + 15.0 I-tm2 was found in the present study. In the present series many of the mean nuclear areas in large cell lymphomas were less than these values. An u n e x p e c t e d finding involved the nuclear characteristics of the small, unstimulated lymphocytes that formed a component: in each of the 20 large cell
lymphomas analyzed morphometrically (4 to 37 per cent o f the total population measured in each case; mean, 17 per cent). These lympocytes have tradiiionally been referred to as residual normal lymphocytes. However, morphometric assessment generally reveals that the small lymphocytes in large cell lymphomas are smaller than normal mantle zone and germinal center small lymphocytes and have irregularities of nuclear shape (NCI) that parallel those of the larger nucleated cells that are usually considered the neoplastic component. T h e exact quantity and topology of nonneoplastic B and T lymphocytes in N H L have not been definitively established. 45-49 T h e current findings suggest that at least some of these small lymphocytes are part o f the t u m o r cell cohort. T h e pathobiologic role and significance of this population remain to be clarified. It is possible tha t they represent part of a similar maturation mechanism used to explain h e t e r o g e n e o u s t u m o r cell populations in N H L o f the follicular center cell type.14A6,50 Despite lack o f understanding of the nature and role of small, unstimulated lymphocytes in large cell l y m p h o m a , their presence is the key to u n d e r standing the spectrum o f nuclear size that occurs in NHL. Both the size and chromatin distribution of the nuclei o f small lymphocytes contrast with the finely distributed, condensed chromatin in the nuclei of the larger tumor cell component of large cell lymphomas. However, it is primarily the size of the nuclei in the small lymphocytes that is responsible for creating the illusion of a "large" nucleated neoplastic cell in the large cell lymphomas in which the mean nuclear area is within the nuclear size range for normal small lymphocytes in mantle zones and germinal centers. Case 2 (table 1) illustrates this phenomenon: the mean nuclear area o f LC 1 (1 t.5 ~m 2) is considerably smaller than that of LC3 (21.2 I.tm2), but the area of the latter is within the range of nuclear area for GC1 lymphocytes shown in figure 2 (16.5 to 24.9 Ixm~). All current classifications o f N H L segregate the various subtypes according to the common denominators of nuclear size and condensed chromatin distribution. However, with the exception of tile recognition o f architectural patterns, current criteria used for categorizing the spectrum of N H L subtypes have been unreliable. 1-5 F u r t h e r m o r e , although it has increased knowledge of the developmental and functional aspects of lymphocytes, the immunologic approach to N H L classification has not enhanced diagnostic capabilities in the morphologic aspects of these disorders.Sl, 52 In terms of patient management, the current morphometric assessment of large cell lymphomas does not simply demonstrate inconsistencies in N H L classification criteria and reasons for int6robserver diagnostic discrepancies. T h e s e results also raise questions concerning the biologic basis for the current classification systems that are thought to reflect the morphologic stages in normal lymphocyte transformation. T h e acc~lracy of the latter hypothesis may be immaterial in view of the findings of the NonH o d g k i n ' s L y m p h o m a Pathologic Classification 970
NUCLEARANALYSESIN NHL [Dardick et al.]
Project. 38 F r o m the p r o g n o s t i c a n d therapeutic viewpoints, with this c l a s s i f i c a t i o n s y s t e m N H L appears to be divided into three b r o a d categories that reflect behavioral aspects, with an a p p a r e n t disregard for the systematic a r r a n g e m e n t o f subtypes in o t h e r classifications. I n light o f the p r e s e n t results indicating t h a t l a r g e cell l y m p h o m a s a r e n o t h o m o g e n e o u s with r e g a r d to n u c l e a r size a n d gross c h r o matin organization, it is possible that nuclear size is not as i m p o r t a n t a prognostic indicator or diagnostic criterion as stressed previously. Clearly, n o t all large cell l y m p h o m a s are truly c o m p o s e d o f neoplastic lymphocytcs with "large" nuclei. More i m p o r t a n t is the visual effect created by the distribution o f c o n d e n s e d c h r o m a t i n . It is obvious that accurate evaluation o f nuclear size in histologic sections by routine observation is difficult. U n f o r t u n a t e l y , visual assessment o f c o n d e n s e d c h r o m a t i n a m o u n t s a n d d i s t r i b u t i o n is also unreliable. 53-56 C l i n i c o p a t h o l o g i c studies o f d i f f u s e histiocytic l y m p h o m a have not s u c c e e d e d in establishing significantly d i f f e r e n t o r c o n s i s t e n t p a t t e r n s o f survival data f o r the various m o r p h o l o g i c subtypes. 4,6,1~,1~,15 However, W a r n k e et al. 4 indicated that it m a y be possible to segregate large cell l y m p h o m a s into two b r o a d prognostic categories. I n s u m m a r i z i n g the literature and their o w n results, these a u t h o r s 4 questioned the" u s e f u l n e s s o f f r a g m e n t i n g d i f f u s e histiocytic lymp h o m a into n u m e r o u s m o r p h o l o g i c subtypes. Because o f the difficulties e n c o u n t e r e d even in distinguishing "small" f r o m "large" nucleated l y m p h o m a s in the c u r r e n t study, this seems to be a reasonable assessment. Perhaps i m p r o v e m e n t s in m o r p h o l o g i c criteria a n d modifications o f seemingly i n a p p r o p r i a t e t e r m i n o l o g y in existing N H L classifications can result f r o m a systematic a n d objective analysis o f neoplastic l y m p h o c y t e s in the various subcategories. Certainly, the biochemical a n d cell biologic aspects o f nuclear organization in n o r m a l a n d l y m p h o m a t o u s l y m p h o cytes must be investigated m o r e t h o r o u g h l y and understood. Results o f clinical trials will be m e a n i n g f u l only w h e n the pathobiologic processes resulting in the diverse s p e c t r u m o f t u m o r o u s lymphocytes are u n d e r s t o o d a n d m e a n i n g f u l a n d r e p r o d u c i b l e cytologic criteria are established.
Acknowledgment. The authors are indebted to the Canadian Pathologists, who contributed case material to tiffs study. REFERENCES 1. Ezdinli EZ, Costell0 W, Wasser LP, et al: Eastern Cooperative Oncology Group experience with the Rappaport classification of non-Hodgkin's lymphomas. Cancer 43:544, 1979 92. Jones SE, Butler JJ, Byrne GE, et al: Histopathologic review of lymphoma cases from the Southwest Oncology Group. Cancer 39:1071, 1977 3. Kim H, Zelman RJ, Fox MA, et al: Pathology panel for lymphoma clinical studies: a comprehensive analysis of cases accumulated since its inception. J Nail Cancer Inst 68:43, 1982 4. Warnke RA, Strauchen JA, Burke ]S, et ah Morphologic types of diffuse large-cell lymphoma. Cancer 50:690, 1982
971
5. Woodruff RD: Reviewing histologic diagnosis in lymphoma. Arch Pathol Lab Med 105:573, 1981 6. Armitage JO, Dick FR, Platz CE, et ah Clinical usefulness and: reproducibility of histologic subclassification of advanced diffuse histiocytic lymphoma. AmJ Med 67:929, 1979 7. Azar HA, Jaffe ES, Berard CW, et al: Diffuse large cell lymphomas (reticulum cell sarcomas, histiocytic lymphomas). Correlation of morphologic features with functional markers. Cancer 46:1428, 1980 8. Berard CW, Jaffe ES, Braylan RC, et al: Immunologic aspects and pathology of tile malignant lymphomas. Cancer 42:911, 1978 9. Lukes RJ, Collins RD: Immunologic characterization of human malignant lymphomas. Cancer 34:1488, 1974 I0. Lukes RJ, Collins RD: New approaches to the classification of the lymphomata. BrJ Cancer 31 (suppl II): 1, 1975 11. LukesRJ, ParkerJW, Taylor CR, et al: Immunologic approach to non-Hodgkin's lymphomas and related leukemias..Analysis of the results of muhiparameter studies of 425 cases. Semin Hematol 15:322, 1978 12. Newcomer LN, Nerenberg MI, Cadman EC, et al: The usefulness of the Lukes-Collins classification in identifying subsets of diffuse histiocytic lymphoma responsive to chemotherapy. Cancer 50:439, 1982 13. Strauchen JA, Young RC, DeVita VT, et ah Clinical relevance of the histopathological subclassification of diffuse "histiocytic" lymphoma. N EnglJ Med 299:1382, 1978 14. Taylor CR: Results of multiparameter studies of B-cell lymphomas. AmJ Clin Pathol 72:687, 1979 15. Whitcomb CC, Cousar JB, Flint A, et al: Subcategories of histiocytic lymphoma: associations with survival and reproducibility of classification. The Southeastern Cancer Study Group experience. Cancer 48:2464, 1981 16. Maurer R, Taylor CR, Parker JW, et al: Immunoblastic sarcoma: morphologic criteria and the distinction of B and T cell types. Oncology 39:42, 1982 17. Abbott CR, Blewitt RW, Bird CC: Quantitative analysis ofnonHodgkin's lymphoma. J Clin Pathol 35:135, 1982 18. Crocker J, Curran RC: A study of nuclear diameters in lymph node imprints using the Zeiss Microvideomat.J Clin Pathol 32:670, 1979 19. Crocker J, Jones EL, Curran RC: Study of nuclear diameters in non-Hodgkin's lympho.mas.J Clin Pathol 35:954, 1982 20. Crocker J, Jones EL, Curran RC: A comparative study of nu9clear form factor, area and diameter in non-Hodgkin's lymphomas and reactive lymph nodes. J Clin Pathol 36:298, 1983 21. EimotoT, Mitsui T, Kikuchi M: Ultrastructure of adult T-cell leukemia/lymphoma. Virchows Arch [Cell Pathol] 38:189, 1981 22. van der Valk P, te Velde J, Jansen J, et al: Malignant lymphoma of true histiocytic origin: a morphological, ultrastructural, cytochemical and clinical study of 10 cases. Virchows Arch [Pathol Anat] 391:249, 1981 23. van der Valk P, Hermans J, Brand R, et al: Morphometric characterization of diffuse large-cell (histiocytic) lymphomas. Am J Pathol 107:327, 1982 24. van der Valk P, Mosch A, Kurver PJ, et ah Morphometric characterisation of 52 B cell non-Hodgkin's lymphomas. J Clin Pathol 36:289, 1983 25. Caorsi I, Figueroa CD, Rodriguez EM: Morphologic and morphometric study of the two main cell lineages involved in mycosis fungoides: the lymphoid cells and Langerhans cells. Uhrastruct Pathol 3:119, 1982 26. Litovitz TL, Lutzner MA: Quantitative measurements of broad lymphocytes from patients with chronic lymphocytic leukemia and the S6zary syndrome. J Natl Cancer Inst 53:75, 1974 27. van der Loo EM, Cornelisse CJ, van Vloten WA, et al: Diagnostic morphometry of isolated lymph node cells from patients with mycosis fungoide s-and S6zary's syndrome. Virchows Arch [Cell Pathol] 33:107, 1980 28. van der Loo EM, van Vloten WA, Cornelisse cJ, et al: The
HUMAN PATHOLOGY
29. 30. 31. 32.
33. 34.
35.
36. 37. 38.
39. 40. 41. 42.
Volume 15, No. 10 (October 1984) 43. Dardick I: Morphometry of normal human lymphoid tissues. Nuclear invaginations and clefts. Arch Pathol Lab Med 108:197, 1984 44. Vollmer RT: Tile effect of cell size on the pathologic diagnosis of small and large cell carcinomas of the lung. Cancer 50:1380, 1982 45. Jaffe ES, Braylan RC, Nanba K, et ah Functional markers: a new perspective on malignant lymphomas. Cancer Treat Rep 61:953, 1977 46. Warnke R, Levy R: Detection of T and B cell antigens with hybridoma monoclonal antibodies: a biotin-avidin-horseradish peroxidase method. J Histochem Cytocllem 28:771, 1980 47. van Heerde P, Fehkamp CA, Fehkamp-Vroom TM, et al: Non-Hodgkin's lymphoma. Immunohistochemicaland electron microscopical findings in relation to light microscopy. A study of 74 cases. Cancer 46:2210, 1980 48. Harris NL, Data RE: The distribution of neoplastic and normal B-lymphoid cells in nodular lymphomas: use of an immunoperoxidase technique on frozen sections. Hv.xl PATHOL 13:610, 1982 49. Dvoretskey P, Wood GS, Lewvy R, et al: T-lymphocyte subsets in follicular lymphomas compared with those in non-neoplastic lymph nodes and tonsils. Hum Pathol 13:618, 1982 50. van den Tweel JG, Lukes RJ, Taylor CR: Pathophysiology of lymphocyte transformation. A study of so-called coutposite lyntphomas. Am J Clin Pathol 71:509, 1979 51. Bain GO: Non-Hodgkin's l)unphomas. Analysis of 92 cases using the "International" classification. Arch Pathol Lab Med 107:64, 1983 52. Jaffe ES, Strauchen JA, Berard CW: Predictability of immunologic phenotype by morphological criteria in diffuse aggressive non-Hodgkin's lymphoma. AmJ Clin Pathol 77:46, 1982 53. Dardick I, Setterfield G: Vohnne of condensed chromatin in developing primitive-lineerythrocytes of chick. Exp Cell Res 100:159, 1976 54. Dardick I, Sheinin R, Setterfield G: Mutant mouse L-cells: a model for megaloblastic anaemia. Br J Haematol 39:483, 1978 55. Setterfield G, Sheinin R, Dardick I, et al: Structure of interphase nuclei in relation to the cell cycle. J Cell Biol 77:246, 1978 56. Sheinin R, Setterfield G, Dardick I, et al: Relationship between chromatin structure and replication in ts mouse L-cells. Can J Biochem 58:1359, 1980
relevance of morphometry in the differential diagnosis of cutaneous T cell lymphomas. BrJ Dermatol 104:257, 1981 Lutzner M, Edelson R, Shein P, et al: Cutaneous T-cell lymphomas: the Sdzary syndrome, mycosis fungoides and related disorders. Ann Intern Med 83:534, 1975 McNutt NS, Heilbron DC, Crain WR: Mycosis fungoides. Diagnostic criteria based on quantitative electron microscopy. Lab Invest 44:466, 1981 McNutt NS, Crain WR: Quantitative electron microscopic comparison of'lymphocyte nuclear contours in mycosis fungoides and inbenign infiltrates in skin. Cancer 47:698, 1981 Meijer CJLM, van der Loo EM, van Vloten WA, et al: Early diagnosis of mycosis fungoides and Sdzary's syndrome by morphometric analysis of lymphoid cells in the skin. Cancer 45:2864, 1980 Scheffer E, Meijer CJLM, van Vloten WA: Dermatopathic lymphadenopathy and lymph node involvement in mycosis fungoides. Cancer 45:137, 1980 Dardick I, Setterfield G, Hall R, et al: Nuclear aherations during l)mphocyte transformation. Relationship to the heterogeneous morphologic presentations of non-Hodgkin's lyntphomas. Am J Pathol 103:10, 1981 Dardick I, Sinnott NM, Hall R, et al: Nuclear morphology and morphometry of B lymphocyte transformation. Implications for follicular center cell lymphomas. Am J Pathol 111:35, 1983 Levine AM, Tayl~or,CR, Schneider DR, et al: Immunoblastic sarcoma of T-cell versus B-cell origin: I. clinical features. Blood 58:52, 1981 Lichtenstein A, Levine AM, Lukes RJ, et al: Inmtunoblastic sarcoma. A clinical description. Cancer 43:343, 1979 Report of the Writing Committee: National Cancer Institute sponsored study of classifications of non-Hodgkin's" lymphomas: summary and description of a working formulation for clinical use. Cancer 49:2112, 1982 Troyer H, Babich E: A hematoxylin and eosin-like stain for glycol methacrylate embedded tissue sections. Stain Technol 56:39, 1981 Schrek R: Ultrastructure of blood lymphocytes from chronic lymphocytic and lymphosarcoma cell leukemia. J Natl Cancer Inst 48:51, 1972 Nathwani BN: A critical analysis of the classifications of nonHodgkin's lymphomas. Cancer 44:347, 1979 Said JW, Hargreaves HK, Pinkus GS: Non-Hodgkin's lymphomas: an ultrastructural stud)' correlating morphology with immunologic cell type. Cancer 44:504, 1979
972