- Email: [email protected]
Impaired natural killer activity in lymphohistiocytosis syndrome
Impaired natural killer activity in lymphohistiocytosis syndrome In six patients with well-documented lymphohistioo'tosis syndrome, natural killer activit)' was found to be profoundly impaired and could not be increased by incubation in vitro with interferon. This abnormality was not found in parents of the affected children..,t clear correlation with the activity of the disease was observed, although a delay of a few weeks (possibly reflecting the life span of NK cells in blood) was seen in the disappearance of NK activity after the onset of the disease and its reappearance after remission. No absolute correlation was observed between N K activity and percentage of leukocytes detected by the Leu-7 monoclonal antibody. Our findings htdicate that testing NK activity is useful for the diagnosis of lyntphohistiocytosis syndrome and can be used as an index of aetiviO' of the disease, among other major clinical and biologic signs of this syndrome. Reversal of the NK activio' defect (rather than detection of Leu-7 positive cells) appears to be a good criterion of complete remission. (J PEmArR I04.'569, 1984)
Nestor Perez, M.D.,* Jean-Louis Virelizier, M.D., Fernando Arenzana-Seisdedos, M.D., Alain Fischer, M.D., and Claude Griscelli, M.D. Paris, F r a n c e LYMPIIOIIISTIOCYTOSIS is a rare syndrome that is frequently familial, with a probable autosomal recessive inheritance, and is usually fatal? --~ Recently, the use of epipodophyllotoxin (VP-16) has resulted in prolonged remission 6 and allowed better evaluation of the biologic abnormalities associated with LHS. For example, minor immunologic abnormalities have been reported in patients
with LHS, but none is found consistently? -9 Nonspecific cytotoxicity mediated by natural killer cells may have a role in host defense against tumors and viruses, ~~is mediated by non-T non-B cells contained in a
small cell population detectable by the Leu-7 monoclonal antibody," and is augmented by interferon. In our experience, the clinical and biologic signs of LtlS resemble those of the so-called "accelerated phase" of Chediak-Higashi ~2 and related syndromes tJ in which NK activity is profoundly impaired ~4"~~and is not reversed by administration of From Unitb d'lmmunologie et d'tt~matologie P~diatriques, attd Groupe de Recherche d'hmmmologie et de Rhuntatologie Pbdiatriques. (hVSERM U 132), ttbpital Necker-Et~ants Malades. Supported in part b)' Fondation pour la Recherche Mbdieale, Paris. *Present address: tlospital de Nihos, LaPlata. Argentina. Reprint requests: dean-Louis Virelizier, M.D.. Unit~ d'Imntunologle et tie Rhumatologie Pkdiatriques, ttbpital Neeker-Enfants Malades. 149 rue de Sbvres, 75743 Paris Cbdex 15, France.
interferon? 6 We report that NK activity is abolished in six patients with LHS, a finding that may have diagnostic and pathogenic implications. METIlODS Before therapy, all six patients had clinical and biologic signs of LHS (Table I). No evidence of recent bacterial or fungal infection could be found. No serologic evidence of recent infection was found for Epstein-Barr virus (antiVCA, EA, or EBNA antibody) in five patients tested, or cytomegalovirus in 6 patients tested. Patients were given VP-16 and steroids. E/T LIIS NK
Effector/target ratio Lymphohistiocytosissyndrome Natural killer
Natural killer activity. This test was performed as described previously? ~ Briefly, before the test effector cells (Ficoll-Hypaque-isolated peripheral mononuclear leukocytes) were incubated for 18 hours in control RPMI medium with 10% human AB serum, or in a preparation of human Sendai-induced leukocyte a-interferon (Institute Pasteur, Paris), at a final concentration of 10,000 IU/ml. Effector leukocytes were washed before co-culture at various effector/target ratios with 5~Cr-labclcd K 562 cells.
TheJournalofPEDIATRICS
569
570
Perez et al.
The Journal of Pediatrics April 1984
T a b l e I. Clinical and laboratory data in six patients with lymphohistiocytosis
Patient
' Sex Siblings dead from lymphohistiocytosis Age at first signs Fever (>38.5~ Hepatosplenomegaly WBC/mm 3 in CSF Blood P M N / m m 3 Platelets/mm 3 Serum fibrinogen (gin/L) (NI 1.8 to 4) Serum triglycerides (mmol/L) (NI 0.55 to 1.70) Erythrophagocytosis Death from LttS
I
2
I
I
4
5
M 0
] F 0
M 0
1 day + 85 400 30,000 0.5
3 wk + + 10 1,470 65,000 0.5
11 mo + + 19 2,200 70,000 0.3
14 mo + + 5 100 20,000 3.0
3 mo + + 150 750 25,000 <1.0
2.1
1.0
4.7
10.6
4.1
3.1
+ +
+
+
+
+ +
F 1
F 1
M 2
+ +
4 mo + + 2 260 65,000 1.25
T a b l e II. N a t u r a l killer activity in six patients with lymphohistiocytosis
Cytotoxic indexes (%) at different E/T ratios after incubation Control medium Patient 1 2 3 4 5 6 Normal values* Adults Children P
I
a-Interferon preparation
50:1
25:1
12:1
6:1
50:1
3 NT <1 6 <1 12
3 6 <1 4
2 6 <1 5 <1 7
2 4 <1 4 <1 I
NT NT <1 9 <1 13
NT NT <1 9 <1 7
NT 4 <1 6 <1 3
NT 2 <1 5 <1 1
52.9 4- 21r 53 ___22 <0.001
41 _ 19 44 • 22 <0.001
27 4- 16 31 4- 24 <0.02
17 4- 13 25 • 20 <0.02
76 4- 15 71 4- 14 <0.001
69 4- 18 64 4- 18 <0.001
544- 18 48 --. 18 <0.001
35+-- 14 34 4- 21 <0.01
125:1112:1]6:1
NT, Not tested. *Results obtained in our laboratory in 39 adults and 10 children. eArithmetic mean • I standard deviation :[Student t test. comparison ~ith control values. Cytotoxic indexes were calculated a f t e r a 4 - h o u r cytolysis assay. Immunofluorescence staining ~ i t h Leu-7 monoclonal antibody. Leu-7 I g M monoclonal a n t i b o d y (clone H N K - 1 ) was purchased from Becton-Dickinson ( R u t h e r f o r d , N J). Five microliters of antibody preparation was added to 1 • 106 leukocytes in 50 #1 phosphate-buffered saline containing 20% bovine serum a l b u m i n . A f t e r 20 minutes' incubation at 4 ~ C, cells were washed t h r e e times with H a n k s medium containing 0.1% sodium azide. T e n microliters of a fluorescein-conjugated goat a n t i - m o u s e i m m u noglobulin ( G A M I G ) ( N o r d i c Laboratories, Tilburg, T h e N e t h e r l a n d s ) was then added in 50 #1 P B S - B S A . T h e cells were incubated for 20 minutes at 4 ~ C a n d washed three times. T h e percentage of fluorescent cells a m o n g mononu-
clear cells was appreciated on a Leitz Dialux 20 microscope. N o staining was observed when the G A M I G was used alone. RESULTS Six unrelated patients with L H S were observed for 2 years. In three patients, a suggestive family history was found, with one or more siblings dead from L H S . T h u s the diagnosis of " f a m i l i a l " histiocytosis could be m a d e in only three cases. N a t u r a l killer activity was profoundly impaired in all six patients observed ( T a b l e II). Spontaneous cytotoxic activity was very low, usually undetectable at all E / T ratios. T h e indexes were thus far beyond those observed in adult controls or children of the same age. Incubation for 18
I/ohtme 104 Number 4 hours in an a-interferon preparation did not modify the cytotoxic indexes at any E / T ratio. This finding was in contrast with the clear enhancement observed in control leukocytes after incubation with interferon (Table II). In three families, parents of the affected child could be tested for NK activity. The results were within normal limits. In three patients, the level of NK activity did not correlate over a period of months with the number of circulating lymphocytes, which remained essentially normal (as in the three other patients in our series) (Figure). In one patient, impairment of NK activity was prolonged for 3 months. During VP-16 therapy, the patient slowly recovered, with disappearance of fever, splenomegaly, and hematologic abnormalities. This was followed by the finding of significant NK activity detected after incubation with interferon in vitro. However, a clinical and biologic relapse was again associated with abolished NK activity. In another patient tested before VP-16 administration, NK activity was profoundly impaired. A first period of VP-16 treatment led to transient clinical improvement, associated with normal NK activity. This was followed by a clear relapse of clinical and biologic signs; at that stage, NK activity was found to be abolished. Remission was again obtained after VP-16 therapy, with parallel reappearance of NK activity. The patient is doing well, without any detectable sign of the disease, and NK activity remains normal with monthly administration of VP-16. In the third patient, NK activity was initially found to be normal. Without treatment, her clinical and biologic condition deteriorated, with secondary appearance of neutropenia and hypofibrinogenemia. At that stage, NK activity was still normal. Two weeks and 4 weeks later, however, NK activity was diminished, but soon returned to normal in parallel with persistent clinical remission. Simultaneous evaluation of NK activity and percentage of Leu-7 positive cells was performed in two patients. In one, a parallel diminution of NK activity and percentage of Leu-7 positive cells was seen following the onset of the syndrome. However, in another patient, NK activity was abolished at a time when Leu-7 positive cells could be clearly detected (data not shown). DISCUSSION Our results show that NK activity is profoundly impaired in patients during the active stage of LHS, whether or not the family history is positive. Thus abnormality of NK activity appears to be an integral part of LHS and should be included among the known biologic signs of this syndrome. In contrast to Chediak-Higashi syndrome, in which the NK defect is permanent, the NK defect of LHS appears to be acquired during the active
Natural killer activity in lymphohistiocytosis
3900 401-2500
/
415 6400
2100 2900 1600 3000
3700 2700 1472 2400 1600 1700
1 m
20.10
i=mBe
m
3.11 8.12.82 28.2
90 2500 1600 4400 3500 1230 1100 2500 1800 1500
A
=
6O
9=-
40
m
6.4
13.4.83
1900 4300 2250 3500
571
I
[PMNm/amm3
PMNImm3 L / r a m :3
.r
2o
~
0
24.5
21.6
21.7 20.10 21.11.82 4.1
7500 3500
280 850
0 1500
5100 18000 1570 4400 2800 2850
14.2
28.2
14.3 30.3 18.5.83
9.3.83 PMN/mm3 Umm3
60 40 20 26.1
Figure. NK activity and absolute counts of lymphocytes (L) and neutrophils (PAIN) during follow-up in three patients. NK cytotoxic indexes after incubation in control medium [] or a-interferon preparation I~. Effector/target ratio was 25:1. At that ratio, control indexes are 41 ___19 without and 69 __. 18 with preincubation with interferon. Arrow, Presence of four or more clinical and biologic abnormalities (see Table I).
phase of the disease. This situation is not exclusive to LHS; the active phase of systemic lupus erythematosus is also associated with an acquired impairment of NK activity, whereas remission is associated with recovery of normal activityJ 7 We could not find any evidence of an active inhibition of NK function. Incubation of LHS leukocytes with indomethacin or of normal leukocytes with LHS serum did not modify NK activity (data not shown), a finding that does not favor an active inhibitory role of prostaglandins or other serum factor on NK activity in LHS. It could be envisaged that hypertriglyceridemia inhibits in vivo the function of NK cells, so that NK activity remains impaired when tested in vitro. This is unlikely, however, because there was no correlation between periods of impairment of NK activity and hypertriglyceridemia, or vice versa, in our patients (data not shown). It is unlikely that VP-16 therapy was responsible for the disappearance of NK activity, for three reasons. First, NK activity was found to be abolished in four patients tested
572
Perez et al.
before the first injections of this drug. Second, N K activity returned to normal despite maintenance of VP-16 therapy. Third, in two patients given VP-16 for myelomonocytic leukemia, N K activity remained normal (unpublished personal observation). Similarly, the impairment of N K function during the active phase of LHS does not result from a lack of activation of N K cells by interferon in vivo, because addition of interferon in vitro could not reverse the defect, whereas addition of interferon does reverse the impairment of N K activity observed in patients with defective endogeneous production of interferon? 6.'3 The striking association between the active phase of LHS and profound impairment of N K activity raises the question of whether the N K defect is a cause or a consequence of this syndrome. The observation that patients with Chediak-Higashi syndrome have a constitutional defect of N K activity and frequently develop an "accelerated phase" (with splenomegaly, fever, neutropenia, thrombopenia, hypertriglyceridemia, hypofibrinogenemia, and erythrophagocytosis) ~2 is compatible with the hypothesis that a preexisting N K defect may have a causal role in the development of this complication. The simplest explanation for our observation that N K activity is abolished in the active phase of L H S and cannot be activated by interferon in vitro is that the NK effector cell is not present in the peripheral blood. Natural killer cells are known to originate from the bone marrow in experimental animals. ~9 In humans, N K activity can be transferred by successful bone marrow transplantation under conditions where 100% of circulating leukocytes are of donor origin in the recipient.:~ In LHS, there is evidence that neutropenia is at least in part a consequence of defective granulopoiesis in bone marrow. It is thus possible to envisage that there is an impaired production of NK effector cells by the bone marrow in LHS. Whatever its pathogenic role in LHS, impaired N K activity appears to be a major biologic sign of this syndrome and may thus be a useful diagnostic criterion. In the very early stage of the disease, however, this sign can still be absent, which is important to underline in order to avoid rejecting the diagnosis of LHS on the finding of normal NK activity at that stage. We have examined infants and children with various infections or immune disorders ~6and never have found a profound defect of N K activity except in diseases easily distinguishable from LHS, such as severe combined immunodefieiency disease or constitutional granulocyte dysfunction, z~ It would be interesting, however, to test N K activity in conditions similar to LHS, such as virus-associated hemophagocytic syndrome. 22 Evaluation of the percentage of Leu-7 positive cells
The Journal of Pediatrics April 1984
appears not to be as good a marker of remission as reversal of N K activity. Only a fraction of Leu-7 positive cells actually effect cytolysis." Thus, the presence of Leu-7 positive ceils does not imply that N K effector cells are circulating in the peripheral blood. Indeed, one of our patients had persistently impaired N K activity with an increasing percentage of Leu-7 positive cells (up to normal levels) during a period of incomplete remission. None of the three patients in this series who eventually died of LHS showed a reversal of N K activity at any time. In contrast, the two patients in whom therapy induced a sustained and complete remission showed a parallel and persistent reversal of the N K defect. In one patient, persistently impaired N K activity was associated with a late relapse, 7 months after the onset of treatment. Altogether, these data indicate that N K activity, rather than the percentage of Leu-7 positive cells, may be a major criterion of complete remission to be evaluated even after disappearance of the other biologic abnormalities of the syndrome. Long-term follow-up of N K activity thus may be useful in monitoring the activity and guiding the therapeutic protocols in patients with LHS. REFERENCES !. Farquhar JW, Macgregor AR, Richmond J: Familial haemophagocytic reticulosis. Br Med J 2:1561, 1958. 2. Nelson P, Santamaria A, Olson RL, Nayak NC: Generalized lymphohistiocytic infiltration. Pediatrics 27:931, 1961. 3. Mozziconacci P, Nezelof C, Attal C, Girard F, Pham HT, Weil J, Desbuquois B, Gadot M: La lymphohistiocytose familiale. Arch Fr Ped!atr 22:385, 1965. 4. McClure PD, Strachan P, Saunders EF: hypofibrinogenemia and thrombocytopenia in familial hemophagocytic reticulosis. J PEDIATR85:67, 1974. 5. Dcvictor D, Fischer A, Mamas S, De Saint Basile G, Durandy A, Buriot D, Griscelli C: Etude immunologique de la lymphohistiocytose familiale. Arch Fr Pediatr 39:135, 1982. 6. Ambruso DR, Hays T, Zwartjes WJ, Tubergen DG, Favara BE: Successful treatment of lymphohistiocytie reticulosis with phagocytosis with epipodophyllotoxin VP-16-213. Cancer 45:2516, 1980. 7. Ladisch S, Holiman B, Poplack DG, Blaese RM: lmmunodeficiency in familial erythrophagocytic lymphohistiocytosis. Lancet 1:581, 1978. 8. Perry MC, ttarrison EG, Burgert EO, Gilchrist GS: Familial erythrophagocytic lymphohistiocytosis Cancer 38:209, 1976. 9. Janka GE, Belohradsky BH, D/iumling S, Miiller-H6cker J, Meister P, Haas R J: Familial lymphohistiocytosis. In Schmalzl F, Huhn D, Schaefer HE, editors: Haematology and blood transfusion. Vol 27: Disorders of lhe monocyte macrophage system. New-York, 1981, Springer-Verlag Berlin Heidelberg pp 245-253. 10. Roder JC, Pross HF: The biology of the human natural killer cell. J Clin Immunol 2:249, 1982. I 1. Abo T, Balch CM: A differentiation antigen of human NK and K cells identified by a monoclonal antibody (lINK-I). J lmmunol 127:1024, 1981. 12. Blume RS, Wolff SM: The Chediak-Higashi syndrome:
Volume 104 Number 4
13.
14.
15.
16.
17.
Studies in four patients and a review of the literature. Medicine 51:247, 1972. Griscelli C, Durandy A, Guy-Grand D, Daguillard F, Herzog C, Pruneiras 3.t: A syndrome associating partial albinism and immunodefieieney. Am J 3.|ed 65:691, 1978. Roder JC, Haliotis T, Klein M, Korec S, Jett JR, Ortaldo J, Heberman RB, Katz P, Fauci AS: A new immunodeficiency disorder in humans involving NK cells. Nature 284:553, 1980. Lipinsky 3,1, Virelizier JL, TurszT, GrisceIli C: Natural killer and killer cell activities in patients with primary immunodeficiencies or defects in immune interferon production. Eur J lmmunol I0:246, 1980. Virelizier JL, Griscelli C: Interferon administration as an immunoregulatory and antimicrobial treatment in children with defective interferon secretion. In Seligmann 3.1, Hitzig WH, editors: Primary immunodeficiencies. Elsevier New York, 1980, North-Holland Biomedical Press, pp 473-484. Nelghbour PA, Grayzel Ai, Miller AE: Endogenous and interferon-augmented natural killer cell activity of human peripheral blood mononuclear cells in vitro: Studies of
Natural killer activity in lymphohistiocytosis
18.
19.
20.
21.
22.
573
patients with multiple sclerosis, systemic lupus erythematosus or rheumatoid arthritis. Clin Exp Immunol 49:11, 1982. Virelizier JL, Griscclli C: D~faut s~lectif de s~crt~tion d'interfcron associ6 fi un d~ficit d'activit~ cytoxique natureIle. Arch Fr Pediatr 38:77, 1981. Hailer O, Kiessling R, Orn A, Wigzell H: Generation of natural killer cells: An autonomous function of the bone marrow. J Exp Med 145:141 i, 1977. Virelizier JL, Lagrue A, Durandy A, Arenzana F, Oury C, Griscelli C, Reinert P: Reversal of natural killer defect in a patient with Chediak-Higashi syndrome after bone marrow transplantation. N Engl J Med 306:1055, 1982. Fischer A, Pham HT, Descamps-Latscha B, Lisowska-Grospierre B, Gerota I, Perez N, Scheinmetzler C, Durandy A, Virclizier JL, Griscelli C: Bone marrow transplantation for inborn error of phagocytic ceils associated with defective adherence, chemotaxis and oxidative response during opsonised particle phagocytosis. Lancet 2:473, 1983. Risdall R J, McKenna RW, Nesbit ME, Krivit W, Balfour tttl, Simmons RL, Brunning RD: Virus-associated hemophagocytic syndrome. Cancer 44:993, 1979.
Nestor Perez, M.D.,* Jean-Louis Virelizier, M.D., Fernando Arenzana-Seisdedos, M.D., Alain Fischer, M.D., and Claude Griscelli, M.D. Paris, F r a n c e LYMPIIOIIISTIOCYTOSIS is a rare syndrome that is frequently familial, with a probable autosomal recessive inheritance, and is usually fatal? --~ Recently, the use of epipodophyllotoxin (VP-16) has resulted in prolonged remission 6 and allowed better evaluation of the biologic abnormalities associated with LHS. For example, minor immunologic abnormalities have been reported in patients
with LHS, but none is found consistently? -9 Nonspecific cytotoxicity mediated by natural killer cells may have a role in host defense against tumors and viruses, ~~is mediated by non-T non-B cells contained in a
small cell population detectable by the Leu-7 monoclonal antibody," and is augmented by interferon. In our experience, the clinical and biologic signs of LtlS resemble those of the so-called "accelerated phase" of Chediak-Higashi ~2 and related syndromes tJ in which NK activity is profoundly impaired ~4"~~and is not reversed by administration of From Unitb d'lmmunologie et d'tt~matologie P~diatriques, attd Groupe de Recherche d'hmmmologie et de Rhuntatologie Pbdiatriques. (hVSERM U 132), ttbpital Necker-Et~ants Malades. Supported in part b)' Fondation pour la Recherche Mbdieale, Paris. *Present address: tlospital de Nihos, LaPlata. Argentina. Reprint requests: dean-Louis Virelizier, M.D.. Unit~ d'Imntunologle et tie Rhumatologie Pkdiatriques, ttbpital Neeker-Enfants Malades. 149 rue de Sbvres, 75743 Paris Cbdex 15, France.
interferon? 6 We report that NK activity is abolished in six patients with LHS, a finding that may have diagnostic and pathogenic implications. METIlODS Before therapy, all six patients had clinical and biologic signs of LHS (Table I). No evidence of recent bacterial or fungal infection could be found. No serologic evidence of recent infection was found for Epstein-Barr virus (antiVCA, EA, or EBNA antibody) in five patients tested, or cytomegalovirus in 6 patients tested. Patients were given VP-16 and steroids. E/T LIIS NK
Effector/target ratio Lymphohistiocytosissyndrome Natural killer
Natural killer activity. This test was performed as described previously? ~ Briefly, before the test effector cells (Ficoll-Hypaque-isolated peripheral mononuclear leukocytes) were incubated for 18 hours in control RPMI medium with 10% human AB serum, or in a preparation of human Sendai-induced leukocyte a-interferon (Institute Pasteur, Paris), at a final concentration of 10,000 IU/ml. Effector leukocytes were washed before co-culture at various effector/target ratios with 5~Cr-labclcd K 562 cells.
TheJournalofPEDIATRICS
569
570
Perez et al.
The Journal of Pediatrics April 1984
T a b l e I. Clinical and laboratory data in six patients with lymphohistiocytosis
Patient
' Sex Siblings dead from lymphohistiocytosis Age at first signs Fever (>38.5~ Hepatosplenomegaly WBC/mm 3 in CSF Blood P M N / m m 3 Platelets/mm 3 Serum fibrinogen (gin/L) (NI 1.8 to 4) Serum triglycerides (mmol/L) (NI 0.55 to 1.70) Erythrophagocytosis Death from LttS
I
2
I
I
4
5
M 0
] F 0
M 0
1 day + 85 400 30,000 0.5
3 wk + + 10 1,470 65,000 0.5
11 mo + + 19 2,200 70,000 0.3
14 mo + + 5 100 20,000 3.0
3 mo + + 150 750 25,000 <1.0
2.1
1.0
4.7
10.6
4.1
3.1
+ +
+
+
+
+ +
F 1
F 1
M 2
+ +
4 mo + + 2 260 65,000 1.25
T a b l e II. N a t u r a l killer activity in six patients with lymphohistiocytosis
Cytotoxic indexes (%) at different E/T ratios after incubation Control medium Patient 1 2 3 4 5 6 Normal values* Adults Children P
I
a-Interferon preparation
50:1
25:1
12:1
6:1
50:1
3 NT <1 6 <1 12
3 6 <1 4
2 6 <1 5 <1 7
2 4 <1 4 <1 I
NT NT <1 9 <1 13
NT NT <1 9 <1 7
NT 4 <1 6 <1 3
NT 2 <1 5 <1 1
52.9 4- 21r 53 ___22 <0.001
41 _ 19 44 • 22 <0.001
27 4- 16 31 4- 24 <0.02
17 4- 13 25 • 20 <0.02
76 4- 15 71 4- 14 <0.001
69 4- 18 64 4- 18 <0.001
544- 18 48 --. 18 <0.001
35+-- 14 34 4- 21 <0.01
125:1112:1]6:1
NT, Not tested. *Results obtained in our laboratory in 39 adults and 10 children. eArithmetic mean • I standard deviation :[Student t test. comparison ~ith control values. Cytotoxic indexes were calculated a f t e r a 4 - h o u r cytolysis assay. Immunofluorescence staining ~ i t h Leu-7 monoclonal antibody. Leu-7 I g M monoclonal a n t i b o d y (clone H N K - 1 ) was purchased from Becton-Dickinson ( R u t h e r f o r d , N J). Five microliters of antibody preparation was added to 1 • 106 leukocytes in 50 #1 phosphate-buffered saline containing 20% bovine serum a l b u m i n . A f t e r 20 minutes' incubation at 4 ~ C, cells were washed t h r e e times with H a n k s medium containing 0.1% sodium azide. T e n microliters of a fluorescein-conjugated goat a n t i - m o u s e i m m u noglobulin ( G A M I G ) ( N o r d i c Laboratories, Tilburg, T h e N e t h e r l a n d s ) was then added in 50 #1 P B S - B S A . T h e cells were incubated for 20 minutes at 4 ~ C a n d washed three times. T h e percentage of fluorescent cells a m o n g mononu-
clear cells was appreciated on a Leitz Dialux 20 microscope. N o staining was observed when the G A M I G was used alone. RESULTS Six unrelated patients with L H S were observed for 2 years. In three patients, a suggestive family history was found, with one or more siblings dead from L H S . T h u s the diagnosis of " f a m i l i a l " histiocytosis could be m a d e in only three cases. N a t u r a l killer activity was profoundly impaired in all six patients observed ( T a b l e II). Spontaneous cytotoxic activity was very low, usually undetectable at all E / T ratios. T h e indexes were thus far beyond those observed in adult controls or children of the same age. Incubation for 18
I/ohtme 104 Number 4 hours in an a-interferon preparation did not modify the cytotoxic indexes at any E / T ratio. This finding was in contrast with the clear enhancement observed in control leukocytes after incubation with interferon (Table II). In three families, parents of the affected child could be tested for NK activity. The results were within normal limits. In three patients, the level of NK activity did not correlate over a period of months with the number of circulating lymphocytes, which remained essentially normal (as in the three other patients in our series) (Figure). In one patient, impairment of NK activity was prolonged for 3 months. During VP-16 therapy, the patient slowly recovered, with disappearance of fever, splenomegaly, and hematologic abnormalities. This was followed by the finding of significant NK activity detected after incubation with interferon in vitro. However, a clinical and biologic relapse was again associated with abolished NK activity. In another patient tested before VP-16 administration, NK activity was profoundly impaired. A first period of VP-16 treatment led to transient clinical improvement, associated with normal NK activity. This was followed by a clear relapse of clinical and biologic signs; at that stage, NK activity was found to be abolished. Remission was again obtained after VP-16 therapy, with parallel reappearance of NK activity. The patient is doing well, without any detectable sign of the disease, and NK activity remains normal with monthly administration of VP-16. In the third patient, NK activity was initially found to be normal. Without treatment, her clinical and biologic condition deteriorated, with secondary appearance of neutropenia and hypofibrinogenemia. At that stage, NK activity was still normal. Two weeks and 4 weeks later, however, NK activity was diminished, but soon returned to normal in parallel with persistent clinical remission. Simultaneous evaluation of NK activity and percentage of Leu-7 positive cells was performed in two patients. In one, a parallel diminution of NK activity and percentage of Leu-7 positive cells was seen following the onset of the syndrome. However, in another patient, NK activity was abolished at a time when Leu-7 positive cells could be clearly detected (data not shown). DISCUSSION Our results show that NK activity is profoundly impaired in patients during the active stage of LHS, whether or not the family history is positive. Thus abnormality of NK activity appears to be an integral part of LHS and should be included among the known biologic signs of this syndrome. In contrast to Chediak-Higashi syndrome, in which the NK defect is permanent, the NK defect of LHS appears to be acquired during the active
Natural killer activity in lymphohistiocytosis
3900 401-2500
/
415 6400
2100 2900 1600 3000
3700 2700 1472 2400 1600 1700
1 m
20.10
i=mBe
m
3.11 8.12.82 28.2
90 2500 1600 4400 3500 1230 1100 2500 1800 1500
A
=
6O
9=-
40
m
6.4
13.4.83
1900 4300 2250 3500
571
I
[PMNm/amm3
PMNImm3 L / r a m :3
.r
2o
~
0
24.5
21.6
21.7 20.10 21.11.82 4.1
7500 3500
280 850
0 1500
5100 18000 1570 4400 2800 2850
14.2
28.2
14.3 30.3 18.5.83
9.3.83 PMN/mm3 Umm3
60 40 20 26.1
Figure. NK activity and absolute counts of lymphocytes (L) and neutrophils (PAIN) during follow-up in three patients. NK cytotoxic indexes after incubation in control medium [] or a-interferon preparation I~. Effector/target ratio was 25:1. At that ratio, control indexes are 41 ___19 without and 69 __. 18 with preincubation with interferon. Arrow, Presence of four or more clinical and biologic abnormalities (see Table I).
phase of the disease. This situation is not exclusive to LHS; the active phase of systemic lupus erythematosus is also associated with an acquired impairment of NK activity, whereas remission is associated with recovery of normal activityJ 7 We could not find any evidence of an active inhibition of NK function. Incubation of LHS leukocytes with indomethacin or of normal leukocytes with LHS serum did not modify NK activity (data not shown), a finding that does not favor an active inhibitory role of prostaglandins or other serum factor on NK activity in LHS. It could be envisaged that hypertriglyceridemia inhibits in vivo the function of NK cells, so that NK activity remains impaired when tested in vitro. This is unlikely, however, because there was no correlation between periods of impairment of NK activity and hypertriglyceridemia, or vice versa, in our patients (data not shown). It is unlikely that VP-16 therapy was responsible for the disappearance of NK activity, for three reasons. First, NK activity was found to be abolished in four patients tested
572
Perez et al.
before the first injections of this drug. Second, N K activity returned to normal despite maintenance of VP-16 therapy. Third, in two patients given VP-16 for myelomonocytic leukemia, N K activity remained normal (unpublished personal observation). Similarly, the impairment of N K function during the active phase of LHS does not result from a lack of activation of N K cells by interferon in vivo, because addition of interferon in vitro could not reverse the defect, whereas addition of interferon does reverse the impairment of N K activity observed in patients with defective endogeneous production of interferon? 6.'3 The striking association between the active phase of LHS and profound impairment of N K activity raises the question of whether the N K defect is a cause or a consequence of this syndrome. The observation that patients with Chediak-Higashi syndrome have a constitutional defect of N K activity and frequently develop an "accelerated phase" (with splenomegaly, fever, neutropenia, thrombopenia, hypertriglyceridemia, hypofibrinogenemia, and erythrophagocytosis) ~2 is compatible with the hypothesis that a preexisting N K defect may have a causal role in the development of this complication. The simplest explanation for our observation that N K activity is abolished in the active phase of L H S and cannot be activated by interferon in vitro is that the NK effector cell is not present in the peripheral blood. Natural killer cells are known to originate from the bone marrow in experimental animals. ~9 In humans, N K activity can be transferred by successful bone marrow transplantation under conditions where 100% of circulating leukocytes are of donor origin in the recipient.:~ In LHS, there is evidence that neutropenia is at least in part a consequence of defective granulopoiesis in bone marrow. It is thus possible to envisage that there is an impaired production of NK effector cells by the bone marrow in LHS. Whatever its pathogenic role in LHS, impaired N K activity appears to be a major biologic sign of this syndrome and may thus be a useful diagnostic criterion. In the very early stage of the disease, however, this sign can still be absent, which is important to underline in order to avoid rejecting the diagnosis of LHS on the finding of normal NK activity at that stage. We have examined infants and children with various infections or immune disorders ~6and never have found a profound defect of N K activity except in diseases easily distinguishable from LHS, such as severe combined immunodefieiency disease or constitutional granulocyte dysfunction, z~ It would be interesting, however, to test N K activity in conditions similar to LHS, such as virus-associated hemophagocytic syndrome. 22 Evaluation of the percentage of Leu-7 positive cells
The Journal of Pediatrics April 1984
appears not to be as good a marker of remission as reversal of N K activity. Only a fraction of Leu-7 positive cells actually effect cytolysis." Thus, the presence of Leu-7 positive ceils does not imply that N K effector cells are circulating in the peripheral blood. Indeed, one of our patients had persistently impaired N K activity with an increasing percentage of Leu-7 positive cells (up to normal levels) during a period of incomplete remission. None of the three patients in this series who eventually died of LHS showed a reversal of N K activity at any time. In contrast, the two patients in whom therapy induced a sustained and complete remission showed a parallel and persistent reversal of the N K defect. In one patient, persistently impaired N K activity was associated with a late relapse, 7 months after the onset of treatment. Altogether, these data indicate that N K activity, rather than the percentage of Leu-7 positive cells, may be a major criterion of complete remission to be evaluated even after disappearance of the other biologic abnormalities of the syndrome. Long-term follow-up of N K activity thus may be useful in monitoring the activity and guiding the therapeutic protocols in patients with LHS. REFERENCES !. Farquhar JW, Macgregor AR, Richmond J: Familial haemophagocytic reticulosis. Br Med J 2:1561, 1958. 2. Nelson P, Santamaria A, Olson RL, Nayak NC: Generalized lymphohistiocytic infiltration. Pediatrics 27:931, 1961. 3. Mozziconacci P, Nezelof C, Attal C, Girard F, Pham HT, Weil J, Desbuquois B, Gadot M: La lymphohistiocytose familiale. Arch Fr Ped!atr 22:385, 1965. 4. McClure PD, Strachan P, Saunders EF: hypofibrinogenemia and thrombocytopenia in familial hemophagocytic reticulosis. J PEDIATR85:67, 1974. 5. Dcvictor D, Fischer A, Mamas S, De Saint Basile G, Durandy A, Buriot D, Griscelli C: Etude immunologique de la lymphohistiocytose familiale. Arch Fr Pediatr 39:135, 1982. 6. Ambruso DR, Hays T, Zwartjes WJ, Tubergen DG, Favara BE: Successful treatment of lymphohistiocytie reticulosis with phagocytosis with epipodophyllotoxin VP-16-213. Cancer 45:2516, 1980. 7. Ladisch S, Holiman B, Poplack DG, Blaese RM: lmmunodeficiency in familial erythrophagocytic lymphohistiocytosis. Lancet 1:581, 1978. 8. Perry MC, ttarrison EG, Burgert EO, Gilchrist GS: Familial erythrophagocytic lymphohistiocytosis Cancer 38:209, 1976. 9. Janka GE, Belohradsky BH, D/iumling S, Miiller-H6cker J, Meister P, Haas R J: Familial lymphohistiocytosis. In Schmalzl F, Huhn D, Schaefer HE, editors: Haematology and blood transfusion. Vol 27: Disorders of lhe monocyte macrophage system. New-York, 1981, Springer-Verlag Berlin Heidelberg pp 245-253. 10. Roder JC, Pross HF: The biology of the human natural killer cell. J Clin Immunol 2:249, 1982. I 1. Abo T, Balch CM: A differentiation antigen of human NK and K cells identified by a monoclonal antibody (lINK-I). J lmmunol 127:1024, 1981. 12. Blume RS, Wolff SM: The Chediak-Higashi syndrome:
Volume 104 Number 4
13.
14.
15.
16.
17.
Studies in four patients and a review of the literature. Medicine 51:247, 1972. Griscelli C, Durandy A, Guy-Grand D, Daguillard F, Herzog C, Pruneiras 3.t: A syndrome associating partial albinism and immunodefieieney. Am J 3.|ed 65:691, 1978. Roder JC, Haliotis T, Klein M, Korec S, Jett JR, Ortaldo J, Heberman RB, Katz P, Fauci AS: A new immunodeficiency disorder in humans involving NK cells. Nature 284:553, 1980. Lipinsky 3,1, Virelizier JL, TurszT, GrisceIli C: Natural killer and killer cell activities in patients with primary immunodeficiencies or defects in immune interferon production. Eur J lmmunol I0:246, 1980. Virelizier JL, Griscelli C: Interferon administration as an immunoregulatory and antimicrobial treatment in children with defective interferon secretion. In Seligmann 3.1, Hitzig WH, editors: Primary immunodeficiencies. Elsevier New York, 1980, North-Holland Biomedical Press, pp 473-484. Nelghbour PA, Grayzel Ai, Miller AE: Endogenous and interferon-augmented natural killer cell activity of human peripheral blood mononuclear cells in vitro: Studies of
Natural killer activity in lymphohistiocytosis
18.
19.
20.
21.
22.
573
patients with multiple sclerosis, systemic lupus erythematosus or rheumatoid arthritis. Clin Exp Immunol 49:11, 1982. Virelizier JL, Griscclli C: D~faut s~lectif de s~crt~tion d'interfcron associ6 fi un d~ficit d'activit~ cytoxique natureIle. Arch Fr Pediatr 38:77, 1981. Hailer O, Kiessling R, Orn A, Wigzell H: Generation of natural killer cells: An autonomous function of the bone marrow. J Exp Med 145:141 i, 1977. Virelizier JL, Lagrue A, Durandy A, Arenzana F, Oury C, Griscelli C, Reinert P: Reversal of natural killer defect in a patient with Chediak-Higashi syndrome after bone marrow transplantation. N Engl J Med 306:1055, 1982. Fischer A, Pham HT, Descamps-Latscha B, Lisowska-Grospierre B, Gerota I, Perez N, Scheinmetzler C, Durandy A, Virclizier JL, Griscelli C: Bone marrow transplantation for inborn error of phagocytic ceils associated with defective adherence, chemotaxis and oxidative response during opsonised particle phagocytosis. Lancet 2:473, 1983. Risdall R J, McKenna RW, Nesbit ME, Krivit W, Balfour tttl, Simmons RL, Brunning RD: Virus-associated hemophagocytic syndrome. Cancer 44:993, 1979.