- Email: [email protected]
Whole blood cell counts and leucocyte differentials in patients with benign or malignant ovarian tumours
European Journal of Obstetrics & Gynecology and Reproductive Biology 72 (1997) 73 77
ELSEVIER
GYliE
Whole blood cell counts and leucocyte differentials in patients with benign or malignant ovarian tumours M . d e n O u d e n a, J . M . H .
U b a c h s a, J . E . G . M .
S t o o t a, J . W . J . v a n W e r s c h b'*
~Department of Gynaecology, De Wever Hospital, P.O. Box 4446, 6401 CX Heerlen, Netherlands bDepartment o,f Haematology, De Wever Hospital, P.O. Box 4446, 6401 CX Heerlen, Netherlamts Received 24 July 1996; revised 21 October 1996" accepted 4 November 1996
Abstract
Objective: The purpose of the study was to determine whether ovarian cancer patients had haematological anomalies compared to patients with benign ovarian tumours. Study design: Whole blood cell counts and leucocyte differentials were measured in 70 patients suspected of having ovarian tumours. Postoperatively, 20 patients had metastatic ovarian cancer and 50 patients had benign ovarian tumours. A control group consisted of 39 healthy women. Cell counting was performed by the Sysmex NE-8000 system. The significance of differences was assessed by the Mann-Whitney U-test or the z2-test, where appropriate. Results: The median haemoglobin and haematocrit levels were significantly lower (P-values 0.004 and 0.02, respectively) and the platelet count was higher (P = 0.02) in the malignant group compared with the benign group. The leucocyte differentials revealed significantly lower values for lymphocytes in the malignant group in comparison to the benign and control groups (P-values 0.02 and 0.00005, respectively). A significant increase in monocytes (P = 0.002) and decreases in eosinophils and basophils (P-values 0.04 and 0.02, respectively) were found in the malignant group compared to the benign group. Conclusion: These results demonstrate significant changes in whole blood cell counts and leucocyte differentials in patients with ovarian cancer. The low lymphocyte counts in the malignant group possibly indicate a primary immunodeficiency as a causal factor. © 1997 Elsevier Science Ireland Ltd. Keywords: Blood cell count; Haematological anomalies; Ovarian cancer
1. Introduction Despite international research for new aetiological and prognostic factors, the mortality rate from advanced ovarian cancer is still high. Differences in blood cell composition between patients with benign and malignant ovarian tumours could provide a better understanding of this disease and consequently improve therapy and prognosis. Moreover, blood cell counts are nowadays readily available. It has been suggested that leucocyte, granulocyte, lymphocyte and monocyte counts may be useful for assessing the prognosis in patients with metastatic gastric
* Corresponding author.
cancer [1]. Lymphocytopenia and monocytosis have been seen in carcinomata of varied origin [2,3]. However, the publications on blood cell alterations in the presence of a malignant tumour are not of recent origin and are based on manual blood cell counting methods. Moreover, most of these studies were performed in heterogeneous patient groups. In the present report the whole blood cell count and the differential white blood cell count were determined with advanced instrumental counting methods in patients with well defined benign or malignant ovarian tumours. The aim was to assess whether haematological anomalies could be found in patients with ovarian malignancies, and if so, to what extent these parameters could provide any differentiation between benign and malignant tumours.
0301-2115/97/$17.00 © 1997 Elsevier Science Ireland Ltd. All rights reserved Pll S0301 - 2 1 1 5 ( 9 6 ) 0 2 6 6 2 - 0
74
M. den Ouden et al. / European Journal o/ Obstetrics & Gynecology amt Reproductive Biology 72 (1997) Z~ 77
Table I Basic patient characteristics of the benign and the malignant tumour patient groups Tumour type
Patients
Benign Serous cystadenoma Mucinous cystadenoma Adenofibroma Endometrial cyst Dermoid cyst
15 16 8 6 5
Malignant Serous adenocarcinoma Mucinous adenocarcinoma Granulosa cell carcinoma
16 2 2
E D T A - K 2 containing tubes (Sarstedt, Germany). The samples were kept at room temperature tbr maximally 5 h before they were run on the Sysmex NE-8000 (Toa Medical Electronics, Kobe, Japan). Reproducibility tests of this system showed a coefficient o f variation o f 2.9'7,, for the neutrophils, 4.2% for the lymphocytes, 8.8% for the monocytes, 8.9% for the eosinophils and 37% for the basophils. The significance o f the differences between the median values o f the various groups was assessed by the M a n n Whitney U-test or the z2-test, where appropriate.
3. Results
2. Materials and methods All patients visiting the outpatient department between October 1992 and J a n u a r y 1995 with an ovarian t u m o u r were entered into the study. Fifty patients (median age 56 years; 25 75th percentile 54 65) appeared to have benign ovarian t u m o u r s with varied histology and twenty patients (median age 61 years; 25 75th percentile 4 6 - 6 5 ) had malignant ovarian tumours at an advanced stage (III/IV) at presentation (Table 1). In the latter group all patients had histologically proven metastases, mostly in lymph nodes or peritoneal tissue. A control g r o u p consisted of 39 agematched subjectively healthy w o m e n (median age, 50 years; 25 75th percentile 45 55). N o n e o f the patients showed clinical signs o f haemostatic abnormalities, vaginal bleeding or thrombosis and their medical histories were free o f t h r o m b o e m b o l i c events. The use o f medication was recorded. Before the operative procedure took place, blood samples were taken between 08.30 and 10.30 after overnight fasting. The samples were drawn into
In Table 2 the results o f the whole blood cell counts of the patients with malignant ovarian turnouts are c o m p a r e d with the values o f the benign ovarian t u m o u r g r o u p and the control group. The group with malignant disease demonstrated significantly lower values for haemoglobin, haematocrit, M C V and M C H in comparison to the benign group. Median values for platelets were higher in the malignant g r o u p in comparison with the benign group. The comparison o f the leucocyte differential is given in Table 3. The malignant group demonstrated significantly lower values for lymphocyte, eosinophil and basophil counts, while m o n o c y t e counts were significantly higher in the malignant group c o m p a r e d to the benign and the control groups. The proportions o f decreased or increased values o f the parameters are listed in Table 4. The percentages o f patients who had decreased lymphocyte counts and haemoglobin levels were significantly higher in the malignant group. In this g r o u p there were significantly more patients with increased neutrophils and m o n o cytes in comparison to the patients with benign tumours.
Table 2 Comparison of the whole blood cell counts of the benign and malignant ovarian cancer patient groups with an age-matched control group Benign ovarian tumours, group 1 (n = 51))
Malignant ovarian tumours. group 2 (n = 20)
Median (25 75th percen- Median (25 75th percentile) tile) Leucocytes (G/l) Haemoglobin (retool/l) Haematocrit (%) MCV (fl) MCH (fmol) MCHC (retool/ 1) Platelets (G/l)
7.75 (5.9 10.0) 8.6 (8.2 9.1) 41.8 (39.7 43.8) 92 (89 96) 1.90 (1.83 1.98) 20.6 (20.4- 20.9) 293 (259 350)
8.1 (7.2 9.5) 8.0 (7.3 8.6) 39.6 (26.8 89 (85 1.81 (I.74 20.2 (19.9
46.3) 91) 1.86) 20.9)
374 (241 493)
Age-matched controls, group 3 0l = 39)
Significance P-value
Median(25 75th percentile)
I vs. 2
1 vs. 3
2 vs. 3
N.S. 0.004
N.S. N.S.
0.01 0.02
0.02 0.04 0.004 N.S.
N.S. N.S. N.S. N.S.
N.S. N.S. 0.01 0.01
0.02
N.S.
N.S.
6.8 (5.5 8.2) 8.6 (8.0 8.9) 40.9 (37.0 90 (87 1.88 (1.82 20.8 (20.4
43.1) 93) 1.94) 21.2)
260 (232 310)
M. den Omh'n
c'{
al. European Join'rod O/ Oh,stct#i~' & GvJtecoloey am/ Repro~hwtice Biology 72 (1997) 7.7 77
75
Table 3 Comparison of the leucocyte differential cotints of ihe benign and malignant ovarian cancer patient groups with an age-matched control group Leucocyte (G/I)
Neutrophils Ly,mphocytes Monocytes Eosinophils Basophils
Benign ovarian tumours. group 1 (n 50)
Malignant ovarian turnouts. group 2 ( n - 2 0 )
Age-matched controls, group 3 01 39)
Signilicance P-value
Median 125 75th percentile)
Median (25 75th percentilel
Median (25 75ih percentile)
t vs. 2
I ~s. 3
2 vs. 3
5.13 (3.80 1.8 (I.3 0.35 (I).26 I).08 (0 0 (0
6.24 (4.61 1.2 (0.7 0.48 (0,43 0 10 0 (0
4.(}1 (3.12 2.0 11.6 0.39 (0.31 0.13 (0.09 0.05 (0.04
N.S. 0.02 0.002 ll.04 0.02
0.02 N.S. N.S. 0.00007 0.01
0.001 0.00005 (I.0(t6 0.00007 0.000008
7.30) 2.31 0.54) 0.13) 0.07)
6.99) 1.7) 0.81) 0.09) 0)
4. Comment
Significant abnormalities in the haematological variables were found at the intake of patients in whom malignant ovarian tumours were subsequently diagnosed. The median haemoglobin and haematocrit levels in both the benign and the malignant groups fell within the normal range, but the percentage of patients having a lower value was significantly higher in the malignani group. There was a significant decrease of MCV and MCH, indicating a trend towards the microcytic hypochromic state and finally resulting in anaemia in the malignant group. Such anaemia is generally mild to moderate and is the most common type of anaemia among hospitalized patients. It is usually called the "anaemia of chronic disease' (ACD) and the three categories of diseases associated with it are infections. inflammations and neoplasia. Changes in the pattern of iron distribution are characteristic of ACD [4]. Significantly higher platelet counts were found in the malignant ovarian tumour group. This finding is in line with the results of Zeimet et al. who found a thrombocytosis in 38% of patients with epithelial ovarian cancer [5]. Reactive thrombocytosis is commonly observed in association with malignant diseases, though its pathophysiology is still poorly understood. It is thought thai humoral factors are responsible for the increase in platelets in ovarian cancer patients. Interleukin-6 (IL-6) was found to be produced by ovarian cancer cell lines [6] and this cytokine has been shown in animal experiments [7] to be able to promote platelet formation. Recently, high I L-6 levels in ascitic fluid have been correlated with thrombocytosis in patients with ovarian cancer [8]. Other factors, such as the macrophage colony-stimulating factor, have been found to be increased in the serum of these patients compared to controls [9]. However, a significant difference in overall survival between patients with and without thrombocytosis has still not been demonstrated. The leucocyte differential in the group with malignant disease revealed a significantly lower lymphocyte count compared to the benign and the control gronps.
4.89) 2.3) 0.48) 0.21) 0.07)
Any explanation for this decrease in lymphocytes remains speculative. In an earlier study lymphopenia appeared to be related to malignancies, but ovarian cancers were not included [2]. Unlike Zbroja-Sontag [10], we found no significant difference m the leucocyte cotints between the benign and the malignant disease groups. Therefore, it might be less likely that the decreased lymphocyte counts in the malignant group in the present study were caused by local inflammatory conditions. As lymphocytes play a major role in the cellular immune mechanism, the question is whether this immunosuppression is caused by the ovarian malignancy, or that immunosuppression and decreased lymphocytes constitute one of the primary causes of ovarian cancer. This theory of a primary immunodeficiency as a cause of malignancy is strengthened by several observations. Malignancy has been found in children with congenital immunodeficiency, in transplant patients maintained on immunosuppressive therapy and in patients after thymectomy. A decreased lymphocyte count is also associated with more aggressive behaviour of the tumour. A recent case-report demonstrated a very aggressively growing malignant ovarian tumour, not responsive to any type of chemotherapy [11], in a H1V positive patient with a dramatic fall in the lymphocyte count.
Though not examined in our study, it is interesting to mention that recently the distribution of activated T lymphocyte subpopulations (CD4 + and CD8 + cells) has been analyzed in patients with epithelial ovarian cancer [12]. More activated T lymphocytes were found in the early stage patients ( l a i b ) , whereas these cells were decreased in the advanced stages. An increased number of activated CD8 + {suppressor) T cells was found in the early stages, suggesting an immunosuppressed state at the early onset of the tumour. Both eosinophil and basophil counts were significantly lower in the malignant disease group patients in comparison to the benign disease and control groups. A decrease in eosinophils or eosinopenia is a rare finding. It can be ascribed to acute infections or to treatment
76
M. den Ouden et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 72 (1997) 73 77
Table 4 Comparison of the proportions of decreased and increased values in the malignant turnout group and of both with the benign tumour group
Neutrophils (g/ l) Lymphocytes (g/l) Monocytes (g/
Patients with benign ovarian turnouts ( n 50)
Patients with malignant ovarian tumours (n = 20)
Decreased values (%)
Increased values (%)
Decreased values (%)
0
48*
0
63*
1.2 7.6
8**
8
37**
5
1.0 3.0
6
8***
0
0
0
0
0 2*****
8**** 2
Laboratory reference range
Increased values (%)
42***
0.14-0.66
b Eosinophils (g/ l) Basophils (g/l) Hb (retool/l)
0
0 15.8"****
5**** 0
0 0.31 0 0.11 7.1 9.9
*P_<0.0001 vs. **P=0.02 vs. ***P<0.0001 vs. ****P<0.0001 vs. *****P= 0.02.
with glucocorticoids, which could not be found in any of our patients. Our finding is in contrast with the increased eosinophil counts in patients with other malignancies [13,14], and might be a phenomenon associated with ovarian cancer. In agreement with our results, Galoppin et al. have demonstrated the combination of low eosinophil and basophil counts in a heterogeneous group of patients with cancer [15]. This led to the hypothesis that common mechanisms might be involved. The low basophil count may be directly due to the presence of the tumour, since the numbers of basophils in the blood have been described as returning to normal after excision of the primary tumour. Finally, in vitro studies have demonstrated that the incidence of tumours and metastases was inversely correlated with the number of blood basophils. The same was found for histamine, produced by the basophil cells [16]. This suggests an anti-tumour activity of basophils and histamine. A low basophil count therefore will have a detrimental effect in patients with cancer. The monocyte count was significantly higher in the malignant group compared to both the benign and the control patient groups. An unexplained increase in blood monocytes might raise the possibility of a malignancy, since 60% of patients with non-haematological malignancy exhibit a monocytosis. Other diseases most frequently causing monocytosis are haematological disorders and vascular collagen diseases. In some other malignancies a correlation has been observed between a higher monocyte count and poorer survival [1]. The prognostic significance of the monocyte count in ovarian cancer is unclear. In the present report the haematological anomalies in patients with metastasized ovarian cancer are demonstrated. The low lymphocyte count is a remarkable finding and may possibly be related to the existence of
a primary immunodeficiency at the start of ovarian cancer. Furthermore, because it is a reliable and relatively cheap parameter to obtain it might be useful for the clinician in diagnosing the disease. However, determination of more specific markers of these and other blood cell components in patients with metastasized and non-metastasized ovarian tumours of different histological types will be needed.
References [l] Bruckner HW, Lavin PT, Plaxe ST, Storch JA, Livstone EM. Absolute granulocyte, lymphocyte, and monocyte counts. JAMA 1982; 47: 1004-1006. [2] Zacharski LR, Linman JW. Lymphocytopenia: its causes and significance. Mayo Clin Proc 1971; 46:168 173. [3] Maldonado JE, Hanlon DG. Monocytosis: a current appraisal. Mayo Clin Proc 1965; 40: 248-259. [4] Sears DA. Anemia of chronic disease. Med Clin North Am 1992; 76: 567--579. [5] Zeimet GA, Marth C, Muller-Holzner E, Daxenbichler G, Dapunt O. Significance of thrombocytosis in patients with epithelial ovarian cancer. Am J Obstet Gynecot 1994; 170:549 554. [6] Whatson JM, Sensintaffar JL, Berek JS, Martinez-Maza O. Constitutive production of interleukin-6 by ovarian cancer cell lines and by primary ovarian tumor cultures. Cancer Res 1990; 50: 6959-6965. [7] Ishibashi T, Kimura H, Shikama T et al. lnterleukine-6 is a potent thrombopoietic factor in vivo in mice. Blood 1989; 74: 1241-1244. [8] Gastl G, Plante M, Finstad CL et al. High IL-6 levels in ascitic fluid correlate with reactive thrombocytosis in patients with epithelial ovarian cancer. Br J Haematol 1993; 83: 433. [9] Xu F J, Ramakrishnan S, Daly L e t al. Increased serum levels of macrophage colony-stimulating factor in ovarian cancer. Am J Obstet Gynecol 1991; 165:1356 1362. [10] Zbroja-Sontag W. Cell-mediated immunity in the blood of women with inflammatory and neoplastic lesions of the ovary. Am J Reprod Immunol 1983; 4: 146-152. [11] Guidozzi F, Reddy S, Wadee A. Characterization of peripheral blood and ovarian cancer-infiltrating lymphocytes: effect of HIV
M. den Ouden et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 72 (1997) 73 77 infection on phenotypic expression and proliferative response. Gynecol Oncol 1994; 53:251 255. [12] Miyazaki K, Shimada K, Katabuchi H, Arakane F, Arao S. Okamura H. Activated (HLA-DR +) T-lympocyte subsets in early epithelial ovarian cancer and malignant ovarian germ cell tumors. Gynecol Oncol 1995; 58:362 367. [13] Kroegel C, Castabel U, Barnes PJ, Matthys H. Die pathogenetische Bedeutung des eosinophilen Granulozyten. II. Hypereosinophile Erkrankungen. Dtsch Med Wochenschr 1988: 113: 1446 1452.
77
[14] Stefanini M, Claustro JC, Motos R, Bendigo LL. Blood and bone marrow eosinophilia in malignant tumors. Cancer 1991~ 68: 543 548. [15] Galoppin L, Noirot C, Wastiaux JP, Scheinmann P, Paupe J, Burtin C. Comparison between basophils, blood histamine, and histamine release in cancer and noncancer patients. J Allergy Clin lmmunol 1989; 84:501 506. [16] Burtin C, Ponvert C, Fray A, et al. Inverse correlation between tumor incidence and tissue histamine levels in W/W ~, W " / + , and ~ + mice. J Natl Cancer Inst 1985; 74: 671.
ELSEVIER
GYliE
Whole blood cell counts and leucocyte differentials in patients with benign or malignant ovarian tumours M . d e n O u d e n a, J . M . H .
U b a c h s a, J . E . G . M .
S t o o t a, J . W . J . v a n W e r s c h b'*
~Department of Gynaecology, De Wever Hospital, P.O. Box 4446, 6401 CX Heerlen, Netherlands bDepartment o,f Haematology, De Wever Hospital, P.O. Box 4446, 6401 CX Heerlen, Netherlamts Received 24 July 1996; revised 21 October 1996" accepted 4 November 1996
Abstract
Objective: The purpose of the study was to determine whether ovarian cancer patients had haematological anomalies compared to patients with benign ovarian tumours. Study design: Whole blood cell counts and leucocyte differentials were measured in 70 patients suspected of having ovarian tumours. Postoperatively, 20 patients had metastatic ovarian cancer and 50 patients had benign ovarian tumours. A control group consisted of 39 healthy women. Cell counting was performed by the Sysmex NE-8000 system. The significance of differences was assessed by the Mann-Whitney U-test or the z2-test, where appropriate. Results: The median haemoglobin and haematocrit levels were significantly lower (P-values 0.004 and 0.02, respectively) and the platelet count was higher (P = 0.02) in the malignant group compared with the benign group. The leucocyte differentials revealed significantly lower values for lymphocytes in the malignant group in comparison to the benign and control groups (P-values 0.02 and 0.00005, respectively). A significant increase in monocytes (P = 0.002) and decreases in eosinophils and basophils (P-values 0.04 and 0.02, respectively) were found in the malignant group compared to the benign group. Conclusion: These results demonstrate significant changes in whole blood cell counts and leucocyte differentials in patients with ovarian cancer. The low lymphocyte counts in the malignant group possibly indicate a primary immunodeficiency as a causal factor. © 1997 Elsevier Science Ireland Ltd. Keywords: Blood cell count; Haematological anomalies; Ovarian cancer
1. Introduction Despite international research for new aetiological and prognostic factors, the mortality rate from advanced ovarian cancer is still high. Differences in blood cell composition between patients with benign and malignant ovarian tumours could provide a better understanding of this disease and consequently improve therapy and prognosis. Moreover, blood cell counts are nowadays readily available. It has been suggested that leucocyte, granulocyte, lymphocyte and monocyte counts may be useful for assessing the prognosis in patients with metastatic gastric
* Corresponding author.
cancer [1]. Lymphocytopenia and monocytosis have been seen in carcinomata of varied origin [2,3]. However, the publications on blood cell alterations in the presence of a malignant tumour are not of recent origin and are based on manual blood cell counting methods. Moreover, most of these studies were performed in heterogeneous patient groups. In the present report the whole blood cell count and the differential white blood cell count were determined with advanced instrumental counting methods in patients with well defined benign or malignant ovarian tumours. The aim was to assess whether haematological anomalies could be found in patients with ovarian malignancies, and if so, to what extent these parameters could provide any differentiation between benign and malignant tumours.
0301-2115/97/$17.00 © 1997 Elsevier Science Ireland Ltd. All rights reserved Pll S0301 - 2 1 1 5 ( 9 6 ) 0 2 6 6 2 - 0
74
M. den Ouden et al. / European Journal o/ Obstetrics & Gynecology amt Reproductive Biology 72 (1997) Z~ 77
Table I Basic patient characteristics of the benign and the malignant tumour patient groups Tumour type
Patients
Benign Serous cystadenoma Mucinous cystadenoma Adenofibroma Endometrial cyst Dermoid cyst
15 16 8 6 5
Malignant Serous adenocarcinoma Mucinous adenocarcinoma Granulosa cell carcinoma
16 2 2
E D T A - K 2 containing tubes (Sarstedt, Germany). The samples were kept at room temperature tbr maximally 5 h before they were run on the Sysmex NE-8000 (Toa Medical Electronics, Kobe, Japan). Reproducibility tests of this system showed a coefficient o f variation o f 2.9'7,, for the neutrophils, 4.2% for the lymphocytes, 8.8% for the monocytes, 8.9% for the eosinophils and 37% for the basophils. The significance o f the differences between the median values o f the various groups was assessed by the M a n n Whitney U-test or the z2-test, where appropriate.
3. Results
2. Materials and methods All patients visiting the outpatient department between October 1992 and J a n u a r y 1995 with an ovarian t u m o u r were entered into the study. Fifty patients (median age 56 years; 25 75th percentile 54 65) appeared to have benign ovarian t u m o u r s with varied histology and twenty patients (median age 61 years; 25 75th percentile 4 6 - 6 5 ) had malignant ovarian tumours at an advanced stage (III/IV) at presentation (Table 1). In the latter group all patients had histologically proven metastases, mostly in lymph nodes or peritoneal tissue. A control g r o u p consisted of 39 agematched subjectively healthy w o m e n (median age, 50 years; 25 75th percentile 45 55). N o n e o f the patients showed clinical signs o f haemostatic abnormalities, vaginal bleeding or thrombosis and their medical histories were free o f t h r o m b o e m b o l i c events. The use o f medication was recorded. Before the operative procedure took place, blood samples were taken between 08.30 and 10.30 after overnight fasting. The samples were drawn into
In Table 2 the results o f the whole blood cell counts of the patients with malignant ovarian turnouts are c o m p a r e d with the values o f the benign ovarian t u m o u r g r o u p and the control group. The group with malignant disease demonstrated significantly lower values for haemoglobin, haematocrit, M C V and M C H in comparison to the benign group. Median values for platelets were higher in the malignant g r o u p in comparison with the benign group. The comparison o f the leucocyte differential is given in Table 3. The malignant group demonstrated significantly lower values for lymphocyte, eosinophil and basophil counts, while m o n o c y t e counts were significantly higher in the malignant group c o m p a r e d to the benign and the control groups. The proportions o f decreased or increased values o f the parameters are listed in Table 4. The percentages o f patients who had decreased lymphocyte counts and haemoglobin levels were significantly higher in the malignant group. In this g r o u p there were significantly more patients with increased neutrophils and m o n o cytes in comparison to the patients with benign tumours.
Table 2 Comparison of the whole blood cell counts of the benign and malignant ovarian cancer patient groups with an age-matched control group Benign ovarian tumours, group 1 (n = 51))
Malignant ovarian tumours. group 2 (n = 20)
Median (25 75th percen- Median (25 75th percentile) tile) Leucocytes (G/l) Haemoglobin (retool/l) Haematocrit (%) MCV (fl) MCH (fmol) MCHC (retool/ 1) Platelets (G/l)
7.75 (5.9 10.0) 8.6 (8.2 9.1) 41.8 (39.7 43.8) 92 (89 96) 1.90 (1.83 1.98) 20.6 (20.4- 20.9) 293 (259 350)
8.1 (7.2 9.5) 8.0 (7.3 8.6) 39.6 (26.8 89 (85 1.81 (I.74 20.2 (19.9
46.3) 91) 1.86) 20.9)
374 (241 493)
Age-matched controls, group 3 0l = 39)
Significance P-value
Median(25 75th percentile)
I vs. 2
1 vs. 3
2 vs. 3
N.S. 0.004
N.S. N.S.
0.01 0.02
0.02 0.04 0.004 N.S.
N.S. N.S. N.S. N.S.
N.S. N.S. 0.01 0.01
0.02
N.S.
N.S.
6.8 (5.5 8.2) 8.6 (8.0 8.9) 40.9 (37.0 90 (87 1.88 (1.82 20.8 (20.4
43.1) 93) 1.94) 21.2)
260 (232 310)
M. den Omh'n
c'{
al. European Join'rod O/ Oh,stct#i~' & GvJtecoloey am/ Repro~hwtice Biology 72 (1997) 7.7 77
75
Table 3 Comparison of the leucocyte differential cotints of ihe benign and malignant ovarian cancer patient groups with an age-matched control group Leucocyte (G/I)
Neutrophils Ly,mphocytes Monocytes Eosinophils Basophils
Benign ovarian tumours. group 1 (n 50)
Malignant ovarian turnouts. group 2 ( n - 2 0 )
Age-matched controls, group 3 01 39)
Signilicance P-value
Median 125 75th percentile)
Median (25 75th percentilel
Median (25 75ih percentile)
t vs. 2
I ~s. 3
2 vs. 3
5.13 (3.80 1.8 (I.3 0.35 (I).26 I).08 (0 0 (0
6.24 (4.61 1.2 (0.7 0.48 (0,43 0 10 0 (0
4.(}1 (3.12 2.0 11.6 0.39 (0.31 0.13 (0.09 0.05 (0.04
N.S. 0.02 0.002 ll.04 0.02
0.02 N.S. N.S. 0.00007 0.01
0.001 0.00005 (I.0(t6 0.00007 0.000008
7.30) 2.31 0.54) 0.13) 0.07)
6.99) 1.7) 0.81) 0.09) 0)
4. Comment
Significant abnormalities in the haematological variables were found at the intake of patients in whom malignant ovarian tumours were subsequently diagnosed. The median haemoglobin and haematocrit levels in both the benign and the malignant groups fell within the normal range, but the percentage of patients having a lower value was significantly higher in the malignani group. There was a significant decrease of MCV and MCH, indicating a trend towards the microcytic hypochromic state and finally resulting in anaemia in the malignant group. Such anaemia is generally mild to moderate and is the most common type of anaemia among hospitalized patients. It is usually called the "anaemia of chronic disease' (ACD) and the three categories of diseases associated with it are infections. inflammations and neoplasia. Changes in the pattern of iron distribution are characteristic of ACD [4]. Significantly higher platelet counts were found in the malignant ovarian tumour group. This finding is in line with the results of Zeimet et al. who found a thrombocytosis in 38% of patients with epithelial ovarian cancer [5]. Reactive thrombocytosis is commonly observed in association with malignant diseases, though its pathophysiology is still poorly understood. It is thought thai humoral factors are responsible for the increase in platelets in ovarian cancer patients. Interleukin-6 (IL-6) was found to be produced by ovarian cancer cell lines [6] and this cytokine has been shown in animal experiments [7] to be able to promote platelet formation. Recently, high I L-6 levels in ascitic fluid have been correlated with thrombocytosis in patients with ovarian cancer [8]. Other factors, such as the macrophage colony-stimulating factor, have been found to be increased in the serum of these patients compared to controls [9]. However, a significant difference in overall survival between patients with and without thrombocytosis has still not been demonstrated. The leucocyte differential in the group with malignant disease revealed a significantly lower lymphocyte count compared to the benign and the control gronps.
4.89) 2.3) 0.48) 0.21) 0.07)
Any explanation for this decrease in lymphocytes remains speculative. In an earlier study lymphopenia appeared to be related to malignancies, but ovarian cancers were not included [2]. Unlike Zbroja-Sontag [10], we found no significant difference m the leucocyte cotints between the benign and the malignant disease groups. Therefore, it might be less likely that the decreased lymphocyte counts in the malignant group in the present study were caused by local inflammatory conditions. As lymphocytes play a major role in the cellular immune mechanism, the question is whether this immunosuppression is caused by the ovarian malignancy, or that immunosuppression and decreased lymphocytes constitute one of the primary causes of ovarian cancer. This theory of a primary immunodeficiency as a cause of malignancy is strengthened by several observations. Malignancy has been found in children with congenital immunodeficiency, in transplant patients maintained on immunosuppressive therapy and in patients after thymectomy. A decreased lymphocyte count is also associated with more aggressive behaviour of the tumour. A recent case-report demonstrated a very aggressively growing malignant ovarian tumour, not responsive to any type of chemotherapy [11], in a H1V positive patient with a dramatic fall in the lymphocyte count.
Though not examined in our study, it is interesting to mention that recently the distribution of activated T lymphocyte subpopulations (CD4 + and CD8 + cells) has been analyzed in patients with epithelial ovarian cancer [12]. More activated T lymphocytes were found in the early stage patients ( l a i b ) , whereas these cells were decreased in the advanced stages. An increased number of activated CD8 + {suppressor) T cells was found in the early stages, suggesting an immunosuppressed state at the early onset of the tumour. Both eosinophil and basophil counts were significantly lower in the malignant disease group patients in comparison to the benign disease and control groups. A decrease in eosinophils or eosinopenia is a rare finding. It can be ascribed to acute infections or to treatment
76
M. den Ouden et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 72 (1997) 73 77
Table 4 Comparison of the proportions of decreased and increased values in the malignant turnout group and of both with the benign tumour group
Neutrophils (g/ l) Lymphocytes (g/l) Monocytes (g/
Patients with benign ovarian turnouts ( n 50)
Patients with malignant ovarian tumours (n = 20)
Decreased values (%)
Increased values (%)
Decreased values (%)
0
48*
0
63*
1.2 7.6
8**
8
37**
5
1.0 3.0
6
8***
0
0
0
0
0 2*****
8**** 2
Laboratory reference range
Increased values (%)
42***
0.14-0.66
b Eosinophils (g/ l) Basophils (g/l) Hb (retool/l)
0
0 15.8"****
5**** 0
0 0.31 0 0.11 7.1 9.9
*P_<0.0001 vs. **P=0.02 vs. ***P<0.0001 vs. ****P<0.0001 vs. *****P= 0.02.
with glucocorticoids, which could not be found in any of our patients. Our finding is in contrast with the increased eosinophil counts in patients with other malignancies [13,14], and might be a phenomenon associated with ovarian cancer. In agreement with our results, Galoppin et al. have demonstrated the combination of low eosinophil and basophil counts in a heterogeneous group of patients with cancer [15]. This led to the hypothesis that common mechanisms might be involved. The low basophil count may be directly due to the presence of the tumour, since the numbers of basophils in the blood have been described as returning to normal after excision of the primary tumour. Finally, in vitro studies have demonstrated that the incidence of tumours and metastases was inversely correlated with the number of blood basophils. The same was found for histamine, produced by the basophil cells [16]. This suggests an anti-tumour activity of basophils and histamine. A low basophil count therefore will have a detrimental effect in patients with cancer. The monocyte count was significantly higher in the malignant group compared to both the benign and the control patient groups. An unexplained increase in blood monocytes might raise the possibility of a malignancy, since 60% of patients with non-haematological malignancy exhibit a monocytosis. Other diseases most frequently causing monocytosis are haematological disorders and vascular collagen diseases. In some other malignancies a correlation has been observed between a higher monocyte count and poorer survival [1]. The prognostic significance of the monocyte count in ovarian cancer is unclear. In the present report the haematological anomalies in patients with metastasized ovarian cancer are demonstrated. The low lymphocyte count is a remarkable finding and may possibly be related to the existence of
a primary immunodeficiency at the start of ovarian cancer. Furthermore, because it is a reliable and relatively cheap parameter to obtain it might be useful for the clinician in diagnosing the disease. However, determination of more specific markers of these and other blood cell components in patients with metastasized and non-metastasized ovarian tumours of different histological types will be needed.
References [l] Bruckner HW, Lavin PT, Plaxe ST, Storch JA, Livstone EM. Absolute granulocyte, lymphocyte, and monocyte counts. JAMA 1982; 47: 1004-1006. [2] Zacharski LR, Linman JW. Lymphocytopenia: its causes and significance. Mayo Clin Proc 1971; 46:168 173. [3] Maldonado JE, Hanlon DG. Monocytosis: a current appraisal. Mayo Clin Proc 1965; 40: 248-259. [4] Sears DA. Anemia of chronic disease. Med Clin North Am 1992; 76: 567--579. [5] Zeimet GA, Marth C, Muller-Holzner E, Daxenbichler G, Dapunt O. Significance of thrombocytosis in patients with epithelial ovarian cancer. Am J Obstet Gynecot 1994; 170:549 554. [6] Whatson JM, Sensintaffar JL, Berek JS, Martinez-Maza O. Constitutive production of interleukin-6 by ovarian cancer cell lines and by primary ovarian tumor cultures. Cancer Res 1990; 50: 6959-6965. [7] Ishibashi T, Kimura H, Shikama T et al. lnterleukine-6 is a potent thrombopoietic factor in vivo in mice. Blood 1989; 74: 1241-1244. [8] Gastl G, Plante M, Finstad CL et al. High IL-6 levels in ascitic fluid correlate with reactive thrombocytosis in patients with epithelial ovarian cancer. Br J Haematol 1993; 83: 433. [9] Xu F J, Ramakrishnan S, Daly L e t al. Increased serum levels of macrophage colony-stimulating factor in ovarian cancer. Am J Obstet Gynecol 1991; 165:1356 1362. [10] Zbroja-Sontag W. Cell-mediated immunity in the blood of women with inflammatory and neoplastic lesions of the ovary. Am J Reprod Immunol 1983; 4: 146-152. [11] Guidozzi F, Reddy S, Wadee A. Characterization of peripheral blood and ovarian cancer-infiltrating lymphocytes: effect of HIV
M. den Ouden et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 72 (1997) 73 77 infection on phenotypic expression and proliferative response. Gynecol Oncol 1994; 53:251 255. [12] Miyazaki K, Shimada K, Katabuchi H, Arakane F, Arao S. Okamura H. Activated (HLA-DR +) T-lympocyte subsets in early epithelial ovarian cancer and malignant ovarian germ cell tumors. Gynecol Oncol 1995; 58:362 367. [13] Kroegel C, Castabel U, Barnes PJ, Matthys H. Die pathogenetische Bedeutung des eosinophilen Granulozyten. II. Hypereosinophile Erkrankungen. Dtsch Med Wochenschr 1988: 113: 1446 1452.
77
[14] Stefanini M, Claustro JC, Motos R, Bendigo LL. Blood and bone marrow eosinophilia in malignant tumors. Cancer 1991~ 68: 543 548. [15] Galoppin L, Noirot C, Wastiaux JP, Scheinmann P, Paupe J, Burtin C. Comparison between basophils, blood histamine, and histamine release in cancer and noncancer patients. J Allergy Clin lmmunol 1989; 84:501 506. [16] Burtin C, Ponvert C, Fray A, et al. Inverse correlation between tumor incidence and tissue histamine levels in W/W ~, W " / + , and ~ + mice. J Natl Cancer Inst 1985; 74: 671.