To the Editor:

When nasal lymphoma is present or a high percentage of granular lymphocytes is detected in the peripheral blood, a natural killer (NK) neoplasm is readily suspected. From the viewpoint of cell-surface antigen analysis, Emile et al1 emphasized the importance of the phenotype, CD7+ CD5 CD2+, as well as the expression of CD56 and the silence of T-cell receptor (TCR) for diagnosis of peripheral-stage NK neoplasm. The phenotype, CD7+ CD5 CD2+, is particularly important in distinguishing NK lymphomas from the peripheral-stage lymphomas vaguely classified as non–B- or T-lineage lymphomas, in which the manifestation of peripheral-stage NK neoplasms such as granular lympohocytes, nasal lymphoma, and angiocentric histology is not overt. The absence of CD5 antigen as well as the expression of CD2 antigen seems to be prerequisite for the diagnosis of peripheral-stage NK neoplasms.

Thus, the phenotypic definitions for the diagnosis of peripheral-stage NK neoplasms appear to be more persuasive than before. On the other hand, the question remaining to be answered is how to distinguish more immature stages of NK neoplasms from cases of acute lymphoblastic leukemia (ALL)/lymphoblastic lymphoma (LBL). The differentiation pathways of T and NK lineages at immature stages are very close,2 and immature NK cells may well be free of mature function or functional molecules such as NK activity or the antigen detected with TIA-1 monoclonal antibody.3 Therefore, it may be very difficult to distinguish T- and NK-lineage neoplasms at the differentiation stages from which ALL/LBL arise. Such neoplasms are not clearly discussed in the Revised European-American Classification of Lymphoid Neoplasms,4 in which much consideration is made as to the derivation site of each neoplasm in the normal hematopoietic differentiation scheme.

In this respect, we also emphasize the phenotype of CD7+ CD5 CD2+ CD56+ in ALL/LBL of NK lineage. In our laboratory, a total of 58 cases of T-lineage ALL/LBL have been phenotyped during the past 9 years.5-8 Nineteen of the 58 cases were CD3+ or CD3 CD4+ CD8+ and of thymic stage. The rest were of early thymic (or prothymic) (CD3 CD4 CD8) stage, including 10 cases of CD7+ CD5+ CD2+, 14 cases of CD7+ CD5+ CD2, 4 cases of CD7+ CD5 CD2+, and 11 cases of CD7+ CD5 CD2. The 11 cases of CD7+ CD5 CD2 included 7 cases with the expression of cytoplasmic CD3ε protein or CD3ε mRNA.6 Three of the 4 CD7+ CD5 CD2+ cases had the RO type expression of CD45 isoformic antigen, in contrast to the RA type that CD7+ CD3 CD4 CD8 (early thymic or prothymic) ALL/LBL generally expressed.8 One of the 3 cases expressed CD56. This CD56+ case has already been published as LBL of probable NK lineage.9 In this case, the entire germline gene configuration of TCR (δ, γ, and β) or Ig, the presence of mRNA of CD3ε, the absence of CD3γ or CD3δ, and the detection of NK-specific dense granules with electron microscopy indicated the NK-lineage derivation. The neoplastic cells had no NK activity. Unexpectedly, the neoplastic cells expressed TdT mRNA as well as TdT protein, as detected by immunofluoresence. Recently, Ichinohazama et al10 also reported a case of LBL of NK lineage with quite the same phenotype that expressed TdT. The expression of TdT is particularly notable, because the gene configuration of Ig or TCR was entirely of germline and, generally, because no information about the expression of TdT is available on physiological NK cells at this stage of differentiation. To our knowledge, these are the only case reports describing NK ALL/LBL.

The low incidence of the phenotype, CD7+ CD5 CD2+ among the CD3 CD4 CD8 cases in our series can be regarded as reflecting the small number of NK cells compared with that of T-lineage cells, whereas the CD7+ CD5 CD2+ (CD3 CD4 CD8) phenotype had formerly been classified into the immature stage of T-lineage cells without solid evidence in the early era of the phenotypic and genotypic analyses of leukemia/lymphoma.11,12 A few cases of CD7+ CD5 CD2+ blasts without monoclonal TCR gene rearrangement can be found in the reports.11,12 Still, the two phenotypes, CD7+ CD5+ CD2 and CD7+ CD5 CD2+, were classified together as CD7+ CD5+/CD2+ immature T-lineage cells in a recent proposal,13 and the CD7+ CD5 CD2+ phenotype were classified simply as an immature stage of T lineage in textbooks,14 15 although there have been no sufficient data on the comparative incidence of those two phenotypes. CD56 is not absolute for NK lineage, because the antigen was expressed in 2 thymic stage cases and a CD7+ CD5+ CD2 case in our results.

Any NK ALL/LBL cells found among the CD7+ CD5 CD2+ (CD56+) ALL/LBL cells should help to clarify the normal differentiation scheme of NK cells at immature stages.

1
Emile
J-F
Boulland
M-L
Haioun
C
Kanavaros
P
Petrella
T
Delfau-Larue
M-H
Bemsussan
A
Farcet
J-P
Gaulard
P
CD5− CD56+ T-cell receptor silent peripheral T-cell lymphomas are natural killer cell lymphoma.
Blood
87
1996
1466
2
Spitz
H
Lanier
LL
Phillips
JH
Development of human T and natural killer cells.
Blood
85
1995
2654
3
Anderson
P
Nagler-Anderson
C
O'Brien
C
Levine
H
Watkins
S
Slayter
HS
Blue
ML
Schlosman
SF
A monoclonal antibody reactive with a 15-kD cytoplasmic granule-associated protein dfines a subpopulation of CD8+ T lymphocytes.
J Immunol
144
1990
574
4
Harris
NH
Jaffe
ES
Banks
PM
Chan
JKC
Cleary
ML
Deisol
G
De Wolf-Peeters
C
Falini
B
Gatter
KC
Grogan
TM
Isaacson
PG
Knowles
DM
Mason
DY
Muller-Hermelink
H-K
Pileri
S-A
Pris
MA
Ralfkiaer
E
Warmkr
RA
A revised European-American classification of lymphoid neoplasms: A proposal from the International Lymphoma Study Group.
Blood
84
1995
1361
5
Tatsumi
E
Yoneda
N
Kawano
S
Yamaguchi
N
CD21 antigen in T-lineage neoplastic lymphoid cells; characteristic expression at thymic stage.
Am J Hematol
45
1994
150
6
Yoneda
N
Tatsumi
E
Teshigawara
K
Nagata
S
Nagano
T
Kishimoto
Y
Kimura
T
Yasunaga
K
Yamaguchi
N
Lineage determination of CD7+ CD5− CD2− and CD7+ CD5+ CD2− lymphoblasts: Studies on phenotype, genotype and gene expression of myeloperoxidase (MPO), CD3ε, and CD3δ chains.
Am J Hematol
45
1994
310
7
Yoneda
N
Tatsumi
E
Kawano
S
Matsuo
Y
Minowada
J
Yamaguchi
N
Recombination activating gene-1 (RAG-1) in leukemia/lymphoma cells: Expression depends on stage of lymphoid differentiation defined by phenotype and genotype.
Blood
82
1993
207
8
Kawano
S
Tatsumi
E
Yoneda
N
Akemi
T
Fumihiko
N
Expression pattern of CD45 RA/RO isoformic antigens in T-lineage neoplasms.
Am J Hematol
49
1995
6
9
Kawano
S
Tatsumi
E
Yoneda
N
Yamaguchi
N
Junko
G
Hiroshi
I
Tomoko
N
Nishikori
M
Okamura
A
Koiwai
O
Novel leukemic lymphoma with probable deviation from immature stage of natural killer (NK) lineage.
Hematol Oncol
13
1995
1
10
Ichinohazama
R
Endoh
K
Ishizawa
K
Okuda
M
Kameoka
J
Meguro
K
Myers
J
Kadin
M
Mori
S
Sawai
T
Thymic lymphoblastic lymphoma of committed natural killer cell precursor origin. A case report.
Cancer
77
1996
2592
11
Van Dongen
JJM
Quertermous
T
Bartram
CR
Gold
DP
Wolves-Tettero
ILM
Commas-Bitter
WM
Hooikaas
H
Adriaansen
HJ
DeKlein
A
Raghavachar
A
Ganser
A
Duby
AD
Seidman
J
Van den Eisen
P
Terhost
C
T-cell receptor-CD3 complex during early T cell differentiation: Analysis of immature T-cell acute lymphoblastic leukemia (T-ALL) at DNA, RNA and cell membrane level.
J Immunol
138
1987
1260
12
Korsmeyer
SJ
Arnold
A
Bakish
A
Ravetch
VJ
Siebenlist
U
Fieter
PA
Sharrow
SO
LeBien
TW
Kersey
JH
Poplack
DG
Leder
P
Waldman
T
Immunoglubulin gene rearrangement and cell surface antigen expression in acute lymphocytic leukemia of T-cell and B-cell precursor origins.
J Clin Invest
71
1983
301
13
Rothe
G
Schmitz
G
Consensus protocol for flow-cytometric immunophenotyping of hematopoietic malignancies.
Leukemia
10
1996
877
14
Rodriguez MA, Cabanillas FC, Deissroth AB: Non-Hodgkin lymphoma, in Gross S, Roath S (eds): Hematology, A Problem-Oriented Approach. Baltimore, MD, Williams and Wilkins, 1996, p 327
15
Hyn BH, Gulati GL, Ashton JK: Laboratory evaluation of acute lymphoblastic leukemia, in Bick RL (ed): Hematology, Clinical and Laboratory Pracitice, vol 2. St Louis, MO, Mosby, 1993, p 795
Sign in via your Institution