Radiotherapy and Oncology 59 (2001) 221226

www.elsevier.com/locate/radonline

Relation of erythrocyte and iron indices to oral cancer growth

Vasudevaru Narayanan Bhattathiri
Clinical Radiobiology Section, Department of Radiotherapy, Regional Cancer Centre, Trivandrum 695-011, India Received 4 May 2000; received in revised form 5 January 2001; accepted 12 February 2001

Abstract Background and purpose: Anaemia is known to inuence prognosis of head and neck cancer patients, but how anaemia and tumour growth inuences each other is not clear. The present study investigates the relation of erythrocyte and iron indices of oral cancer patients to primary tumour size (Tsize), invasiveness and lymph node involvement. Materials and methods: The haemoglobin (Hb), erythrocyte count (RBC), packed cell volume (PCV), mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH), mean corpuscular haemoglobin concentration (MCHC), Serum iron (SFe), transferrin iron-binding capacity (TIBC) and transferrin saturation (%Fe) were evaluated in 217 untreated patients with epidermoid cancer of the bucco-gingivopalatine area. The association of erythrocyte and iron indices with sex, tumour size groups, invasion of adjacent structures and lymph node involvement, as well as the relation of SFe to Hb were analyzed. Results: Most of the patients were anaemic in terms of Hb (63%), RBC (43%) and PCV (48.4%) but almost all had normal or higher MCH (97.3%) and MCV (93.3%) though MCHC was less than normal in 70.7%. Normal or higher SFe was seen in nearly 70% and TIBC in 45% of patients. Hb, RBC and PCV were signicantly lower in women, but there was no difference between men and women in the case of MCV, MCH and MCHC. Primary tumour size showed negative association with Hb, RBC and PCV but positive association with MCH ,2 cm: 29.7 pg; 24 cm: 31.4 pg; . 4 cm: 31. 7 pg; P 0:04) and MCHC ,2 cm: 29.9; 24 cm: 31.5; . 4 cm: 32.1; P 0:006). MCV, SFe, TIBC and %Fe did not show any relation to primary tumour size. None of the indices had any relation to invasion of adjacent structures or lymph node involvement. MCH, MCHC and MCV were not different in men and women but women had signicantly lower Hb, RBC and PCV. The SFe showed poor correlation with Hb. Conclusions: The negative association of Hb, RBC and PCV with tumour size is most likely due to chronic RBC destruction, probably tumour induced, with the products of haemolysis such as polyamines, glutathione, iron, etc., promoting tumour growth, and the positive association with MCH and MCHC reects compensatory regeneration attempts by bone marrow. Lack of relation between the iron indices and tumour parameters and the poor correlation between SFe and Hb is probably due to utilization of iron by both bone marrow and tumours. Lack of difference in MCH and MCHC between men and women obviates the need of using separate cut-off values for the two sexes, unlike Hb, RBC and PCV. The study suggests that anaemia in oral cancer patients represents a tumourhost interaction and that evaluation of all erythrocyte indices should be part of research on cancer related anaemia. q 2001 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Mean corpuscular hemoglobin; Mean corpuscular hemoglobin concentration; Erythrocyte indices; Iron; Anemia; Oral cancer

1. Introduction The measurement of haemoglobin (Hb) in cancer patients is given importance primarily because it is considered as a surrogate for tumour oxygenation. But there are variations in what is accepted as normal Hb level, with different cut-off values being used in various studies and for men and women [8,15,19,36,48,50]. Tumours in anaemic hosts have higher incidence of metastasis [37,48]. Even though Hb is a surrogate for tissue oxygenation, lack of correlation of pO2 to anaemia is reported [24]. Thus, the inter-relationship of anaemia and tumour are not clear. Iron has a well-dened role in cellular proliferation and Hb synthesis, yet the relation of iron with clinical tumour progression is not well

documented. Iron deciency is common in this part of the world, cancer patients not exempted. All the above reasons prompted the present study exploring the relation of various erythrocyte and iron indices with tumour size, invasiveness and lymph node involvement. 2. Materials and methods The Regional Cancer Centre, Trivandrum, caters to the oncological needs of most of Kerala and other parts of South India. The patients are rst seen in the radiotherapy outpatient, in two clinical units. The author has been seeing most of the head and neck cancer patients registered in the rst unit. Since 1991, more than 2000 patients with epidermoid

0167-8140/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0167-814 0(01)00326-7

222

V.N. Bhattathiri / Radiotherapy and Oncology 59 (2001) 221226

carcinomas of various intra-oral subsites have been seen and evaluated by the author to identify the various tumour, host and treatment factors that inuence tumour aggressiveness and radiation response as well as acute normal tissue reactions, results of some of which have been published earlier [35]. From among these patients, 217 with cancers of the bucco-gingivo-hard palate area in whom the relevant investigations (vide infra) were available, only are included in the present analysis. The other patients were excluded either because they had tumours of tongue and oor of mouth which behave differently [6] or because the investigations could not be done in them. All the patients were seen and evaluated by one person, the author. Primary tumour size was directly measured in the patient. Invasion and lymph node involvement was evaluated by clinical examination, with addition of radiography to evaluate bone involvement. Table 1 details the characteristics of the patient population. The following erythrocyte indices were evaluated: Hb, red blood cell (RBC) count, packed cell volume (PCV), mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH) and mean corpuscular haemoglobin concentration (MCHC) using a semi-automatic haematological counter (Miles India Ltd.) in all the patients. Serum iron (SFe), transferrin iron-binding capacity (TIBC) and percentage transferrin iron saturation (%Fe) were evaluated by bathophenanthrolene method using kits (Merck). The iron markers were available only in 113 patients. Categorization on the basis of sex, presence or absence of LN involvement, and invasion of adjacent tissues was done, and the relation to the indices were analyzed by Mann Whitney U-test. The tumours were categorized on the basis primary tumour size to , 2 cm, 24 cm and . 4 cm, and the relation to the indices analyzed by Kruskal Wallis test. The correlation between SFe and Hb was also explored by calculating the Pearson's correlation coefcient r. This was done for all the patients together and also after sub-grouping according to sex and primary tumour size. Since a large number of comparisons were made, only P , 0:01 was considered as signicant. Analysis was done using Epi-Info (version 5.0; Centers for Disease Control, USA) and GraphPad Prism (version 2.01; GraphPad Software Inc., USA) software.

normal or higher in more than 76.2% men and 60% of women. More than half of the men and women had lower than normal TIBC. The details are given in Table 2. 3.2. Relation to sex The mean Hb was signicantly higher in men as compared to women (12. 0 vs. 11.2; P , 0:00001). This was true for RBC count (3.8 vs. 3.6; P 0:001) and PCV (mean 0.38 vs. 0.36; P 0:002) too. Unlike Hb, RBC and PCV, there was no signicant difference between the two sexes in the case of MCH, MCHC and MCV. Men had signicantly higher SFe than women (72.5 vs. 92.8; P 0:01 3). There was no signicant difference between men and women in the case of TIBC or %Fe. The details are given in Table 3. 3.3. Relation to tumour size Hb, RBC and PCV count showed decrease with increasing tumour size which was nearly signicant in the case of RBC count (P 0:01) and PCV (P 0:01) in women. When all patients were considered together, MCHC showed signicant increase with increasing tumour size, which was signicant at the 0.05 level in the case of MCH. When men and women were considered separately the increase was not signicant. MCV did not have any relation to tumour size. None of the iron indices had any relation with tumour size. The details of these are given in Table 4.
Table 1 Patient and tumour characteristics Total number Sex Men Women Age (years) Range Mean Sites Buccal mucosa Gingiva Hard palate Combined Primary tumour size ,2 cm 24 cm .4 cm Invasion Absent Present Skin Bone Both Trismus Lymph node involvement Present Absent 217 125 92 3482 59.4 55 16 5 141 18 123 76 150 67 38 26 2 1 23 194

3. Results 3.1. Relation to normal reference values When compared with reference values, available from Das [10], in men the Hb, RBC count and PCV were less than normal in 76.8, 52 and 59.3%, respectively. In the case of women, these were 40.2, 31.1 and 33.3%. MCHC was lower than normal in most of the patients (65.9% of men and 77.5% of women), but MCH was higher than normal in 28.5% of men and 21.1% of women, and MCV was higher in most of the men and women (93.5 and 93.3%, respectively). SFe was

V.N. Bhattathiri / Radiotherapy and Oncology 59 (2001) 221226 Table 2 Distribution of RBC and iron indices in relation to normal values (mean (SD)) a Parameter Patient group Normal Proportion (%) of patients with values ,Normal Hb (mg/dl) RBC ( 10 12/l) PCV (%) MCH (pg) MCHC (g/dl) MCV () Serum Fe (mg/100 ml) TIBC (mg/100 ml) All Men Women All Men Women All Men Women All Men Women All Men Women All Men Women All Men Women All Men Women 1316 1114.5 46.6 3.54.5 4054 3547 2732 2732 2732 3236 3236 3236 7694 7694 7694 60160 60160 60160 280400 280400 280400 63 77 40 43 52 31 48 59 33 2 1 4 71 66 78 1 1 0 33 29 40.0 55 55 56 Normal 37 23 60 55 48 65 51 41 66 72 71 75 23 27 18 7 6 7 61 60 60.0 39 41 35 .Normal 0 0 0 2 0 4 1 0 1 26 28 21 6 7 4 93 93 93 6 11 0 6 4 9

223

a RBC, red blood cells; Hb, haemoglobin; PCV, packed cell volume; MCH, mean corpuscular haemoglobin; MCHC, mean corpuscular haemoglobin concentration; MCV, mean corpuscular volume; TIBC, transferrin iron-binding capacity.

3.4. Relation to invasion and lymph node metastasis Subgroup analysis of each of the erythrocyte or iron indices according to presence or absence of invasion and LN metastasis showed no signicant relation for any of the indices. 3.5. Correlation between Hb and Fe When all patients were considered together it was seen that the correlation between SFe and Hb was low (r 0:27), even though statistically signicant (P , 0:005). When analyzed separately for men and women, it was not signicant in the case of men, but signicant at the 0.05 level in the case of women (P , 0:02). The correlation between SFe and Hb was maximum in patients with tumour size , 2 cm (r 0:65) though statistically not signicant, less but signicant (r 0:39; P , 0:005) in the 24 cm size group, and least, but not signicant in the . 4 cm group. 4. Discussion In the present study, nearly two-thirds of the patients were anaemic in terms of Hb reference values. Anaemia in an untreated cancer patient occurs due to a variety of reasons, related or unrelated to the malignancy, such as defective marrow function due to impaired production of erythropoie-

tin, tumor encroachment, myelobrosis or marrow necrosis, low iron availability due to nutritional deciency or enhanced retention of iron in the reticuloendothelial system, increased RBC loss due to haemorrhage or haemolysis, or as a paraneoplastic syndrome [25,30,34,42,46]. In the present study the frequency of patients with low erythropoietin levels is unknown. But the normal or higher than normal MCH in almost all patients, which indicates that active marrow regeneration is taking place, precludes low erythropoietin level as a major cause. It may be argued that this high frequency of anaemia could be due to the greater nutritional deciency, especially that of iron and folic acid, that patients with larger tumours are prone to have. The higher
Table 3 Relation of RBC and iron indices to sex (mean (SD)) Parameter Hb (g/dl) RBC ( 10 12/l) PCV (%) MCV () MCH (pg) MCHC (g/dl) Serum Fe (mg/100 ml) TIBC (mg/100 ml) %Fe
a

Women 11.2 (1.2) 3.6 (0.5) 0.36 (0.05) 99.7 (4.8) 31.0 (2.9) 31.1 (2.6) 73.5 (34.6) 275.5 (86.9) 28.6 (15.1)

Men 12.0 (1.3) 3.8 (0.5) 0.38 (0.05) 99.4 (5.5) 31.6 (2.9) 31.9(2.9) 92.8 (43.7) 271.0 (7.8) 34.8 (14.7)

P value a ,0.00001 0.001 0.002 n.s. n.s. n.s. 0.013 n.s. ,0.05

n.s., not signicant.

224

V.N. Bhattathiri / Radiotherapy and Oncology 59 (2001) 221226

Table 4 Relation of RBC and iron indices to primary tumour size (mean (SD)) Parameter Hb (g/dl) RBC (x10 12/l) PCV (%) MCV () MCH (pg) MCHC (g/dl) Serum Fe (mg/100 ml) TIBC (mg/100 ml) %Fe Group All Women Men All Women Men All Women Men All Women Men All Women Men All Women Men All Men Women All Men Women All Men Women ,2 cm 11.9 (1.3) 12.0 (1.3) 11.7 (1.3) 4.0 (0.6) 4.1 (0.7) 3.9 (0.4) 0.40 (0.06) 0.41 (0.07) 0.39 (0.04) 99.5 (2.9) 99.2 (3.5) 100.0 (2.0) 29.7 (1.5) 29.6 (1.9) 29.9 (0.5) 29.9 (1.5) 29.9 (1.9) 29.9 (0.6) 72.0 (57.4) 105.0 (80.6) 55.5 (47.3) 269.5 (45.3) 299.0 (2.8) 254.8 (50.4) 25.4 (18.1) 35.0 (26.6) 20.7 (14.8) 24 cm 11.7 (1.3) 11.2 (1.1) 12.1 (1.4) 3.8 (0.5) 3.7 (0.4) 3.8 (0.5) 0.37 (0.04) 0.37 (0.04) 0.38 (0.04) 99.6 (4.4) 100.0 (5.4) 99.3 (8.7) 31.4 (2.9) 30.8 (2.6) 31.8 (3.0) 31.5 (2.4) 30.8 (1.8) 32.0 (2.6) 88.2 (38.2) 96.7 (43.4) 76.9 (33.5) 275.4 (82.3) 282.0 (80.3) 264.4 (86.6) 34.5 (16.3) 36.0 (16.7) 32.1 (16.8) .4 cm 11.5 (1.3) 11.0 (1.2) 11.8 (1.2) 3.7 (0.5) 3.5 (0.4) 3.8 (0.5) 0.36 (0.05) 0.34 (0.34) 0.37 (0.05) 99.5 (6.8) 100.0 (4.2) 100.0 (8.0) 31.7 (2.9) 32.0 (3.4) 31.5 (2.5) 32.0 (3.2) 32.2 (3.7) 31.8 (2.9) 81.8 (41.8) 87.6 (43.0) 73.0 (34.0) 270.5 (76.8) 257.5 (62.6) 291.6 (93.6) 30.9 (13.0) 33.5 (12.7) 26.6 (12.7) P value n.s. n.s. n.s. n.s. 0.01 n.s. ,0.05 0.01 n.s. n.s. n.s. n.s. 0.02 n.s. n.s. 0.006 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

than normal MCV (which could indicate folate deciency) in a large number of patients also suggests this. But the normal or higher than normal SFe in most of the patients rules out iron deciency as a major cause of the anaemia. Also, nutritional deciency does not explain the association of MCH and MCHC with tumour size. There are many reports on haemolysis in cancer bearing hosts. Decreased RBC deformability [2,7,11,25,28], presence of autoantibodies to erythrocytes [44], microangiopathic haemolytic anaemia [35], etc., are the common causes of haemolysis. Quemener et al. [39] found that in tumour-bearing rats there was reduction in erythrocytes, Hb and plasma iron which correlated with tumour size, and suggested haemolysis due to alterations in RBC rheological properties as the cause. The products of haemolysis in turn can promote tumour growth. Quemener et at. [39] had noticed splenomegaly in tumour-bearing rats and the tumours in such animals showed decreased growth after splenomegaly, and suggested the release of high concentrations of polyamines from lysed erythrocytes as the reason. In humans polyamines such as N8-acetytspermidine and cadaverine are higher in oral cancer tissue compared to benign tumours and normal tissue [16]. The most probable explanation for the decrease in Hb, RBC count and PCV observed in this study is the presence of tumour induced haemolysis which in turn is related to the advanced/aggressive nature of the disease. The products of haemolysis can

promote tumour growth and contribute to the association. The positive association of tumour size with MCH and MCHC probably reects the rapid compensatory regeneration attempts by the marrow consequent to haemolysis and anaemia. Nearly 70% of the body iron content is in Hb. Iron deciency, as evidenced by a low SFe, leads to lowered Hb. However, iron has a role not only in synthesis of Hb but also in cellular growth, including tumour development and progression. Iron excess aids tumour development by catalyzing the production of oxygen radicals which may be proximate carcinogens and by being a limiting nutrient to the growth and replication of a cancer cell [21,47]. Tumour growth is enhanced by iron as observed by cell culture [22], animal [32,43], and human studies [9,14,31,52]. Many tumours take up iron avidly as evidenced by the higher than normal tissue iron content in many pre-malignant and malignant tissues [12,13,20,38]. Tumour cells are known to overexpress transferrin receptors compared to normal cells and benign tumours [17,27,29,33,45,49], and transferrin receptor expression is correlated with rapid tumour growth rate, poor histology and poor prognosis [1,42,53,55]. Increased iron utilization by tumours can lead to hypotransferrinaemia, decreased iron availability to bone marrow and anaemia. It is suggested that tumour and bone marrow competes with each other for iron [17]. Patients with such tumours can have low Hb even in presence of high SFe. The

V.N. Bhattathiri / Radiotherapy and Oncology 59 (2001) 221226

225

dual role of iron in Hb synthesis and cellular growth and the competition between tumour and bone marrow explains the poor correlation between SFe and Hb, and the lack of any relation of the iron markers, SFe, TIBC and %Fe, with tumour size. It is noteworthy that patients with larger tumours, which are more likely to utilize iron, had least correlation between SFe and Hb, and those with small tumours, which are least likely to utilize large quantities of iron, had highest. Around half the patients in the present study had normal or higher than normal TIBC. High TIBC could be a response to iron deciency. Lack of relation of TIBC with any of the tumour parameters is again due to the complex role that iron and transferrin has in Hb synthesis and tumour growth. It is reported that tumours in anaemic hosts have higher incidence of metastasis [37,48]. In the present study, none of the indices was related to lymph node metastasis. Invasiveness also did not appear to be related to any of the indices. Among the RBC indices Hb, RBC and PCV were significantly different between the two sexes, whereas MCH, MCHC and MCV were not. A study of normal adult reference level Coulter `S' haematological indices also reported signicant sex-related differences for all measurements other than MCV and MCH [40]. Use of MCH, MCHC and MCV may obviate the need for separate cut-off values for men and women. Tobacco abuse, the common culprit for oral cancers, can lead to alterations in iron metabolism and Hb synthesis [18,51]. Studies in animals [26] and humans [23,41,54] show that alcohol ingestion can decrease erythrocyte count, PCV and Hb, and elevate MCV and MCH. These, as well as variations in iron and vitamin intake, could have inuenced the indices in the patients in the present study. But, in spite of their possible presence, the present relation to tumour size is best explained by tumour-induced haemolysis. The present study suggests that, even in head and neck cancer patients, determination of all the indices, rather than Hb alone, is appropriate to evaluate anaemia.. This is likely to give better insight to the cause and prognostic effects of anaemia, and be helpful in discriminating between different types of anaemia and selecting the appropriate management. For example, the prognosis of a patient with anaemia induced by a rapidly growing tumour which avidly utilizes iron or induces haemolysis might be quite different from that of one with anaemia due to iron deciency or low erythropoietin levels. Analysis of patients from the group included in the present paper who were given radical radiotherapy showed that Hb did not inuence primary tumour control whereas MCH and MCHC showed positive, and SFe showed negative inuence. Patients having large tumours associated with rapid growth rate and haemolysis are more likely to have lower Hb than those with large but slowgrowing tumours which do not induce haemolysis. This may be an added reason for the greater prognostic inuence of Hb in larger tumours [36].

Acknowledgements This study was funded by a grant (G. O. MS146/1991) to the author from the Science, Technology and Environment Department, Government of Kerala. References
[1] Basar I, Ayhan A, Bircan K, Ergen A, Tasar C. Transferrin receptor activity as a marker in transitional cell carcinoma of the bladder. Br J Urol 1991;67:165168. [2] Bernat SI, Pongracz E. Hemorrheologic changes in cancer patients. Orv Hetil 1994;135:961964. [3] Bhattathiri VN, Sreelekha TT, Sebastian P, Remani P, Chandini R, Vijayakumar T, Nair MK. Inuence of plasma GSH level on acute radiation mucositis of the oral cavity. Int J Radiat Oncol Biol Phys 1994;29:383386. [4] Bhattathiri VN, Bindu L, Remani P, Chandralekha B, Davis CA. Krishnan Nair Serial cytological assay of micronucleus induction a new tool to predict human cancer radiosensitivity. Radiother Oncol 1997;41:139142. [5] Bhattathiri VN, Bharathykkutty C, Prathapan R, Chirayathmanjiyil DA, Nair KM. Prediction of radiosensitivity of oral cancers by serial cytological assay of nuclear changes. Radiother Oncol 1998;49(1):6165. [6] Borges AM, Shrikhande SS, Ganesh B. Surgical pathology of squamous carcinoma of the oral cavity: its impact on management. Semin Surg Oncol 1989;5:310317. [7] Carla G, Porzio G, Pattizio G, Pietroletti R, Giandomenico G, Amicucci G. Use of an animal model in the study of changes in erythrocyte deformability caused by breast cancer. Minerva Ginecol 1991(43):327329. [8] Chatani M, Matayoshi Y, Masdaki N, Teshima T, Inoue T. Long term follow-up results of high-dose rate remote after loading intracavitary radiation therapy for carcinoma of the uterine cervix. Strahlenther Oncol 1994;170:269276. [9] Damber L, Larsson LG. Combined effects of mining and smoking in the causation of lung carcinoma. A case-control study in northern Sweden. Acta Radiol Oncol 1982;21:305313. [10] Das KVK. Hematology: general considerations. In: Das KVK, editor. A short textbook of medicine, New Delhi: Jaypee Brothers Medical, 1990. pp. 562575. [11] Deliconstantinos G, Villiotou V, Stavrides JC, Salemes N, Gogas J. Nitric oxide and peroxynitrite production by human erythrocytes: a causative factor of toxic anemia in breast cancer patients. Anticancer Res 1995;15:14351446. [12] Dumont AE, Ford RJ, Becker FE. Siderosis of lymph nodes in patients with Hodgkin's disease. Cancer 1976(38):12471252. [13] Durak I, Kavutcu M, Canbolat O, Isik AU, Akyol O. Concentrations of some major and minor elements in larynx tissues with and without cancer. Biometals 1994;7:4548. [14] Edling C. Lung cancer and smoking in a group of iron ore miners. Am J Ind Med 1982;3:191199. [15] Ayes AW, Pintilie M, Kirkbride P, Levin W, Manchul LA, Rawlings GA. Prognostic factors in patients with cervix cancer treated by radiation therapy: results of a multiple regression analysis. Radiother Oncol 1995;35:107117. [16] Gallesio C, Colombatto S, Modica R. Free and acetylated polyamines as markers of oral cavity tumors. Oral Surg Oral Med Oral Pathol 1994;77:167171. [17] Gatter KC, Brown G, Trowbridge IS, Woolston RE, Mason DY. Transferrin receptors in human tissues: their distribution and possible clinical relevance. J Clin Pathol 1983;36:539545. [18] Gavalov SM, Soboleva MK, Deriagina LP, Demchenko AE. Effect of active and passive smoking on the course of pregnancy in women and

226

V.N. Bhattathiri / Radiotherapy and Oncology 59 (2001) 221226 on the establishment of the erythrocytic system in their children. Ter Arkh 1991;63(3):126130. Ginnsky T, Rey A, Roche B, et al. Overall treatment time in advanced cervical carcinomas: a critical parameter in treatment outcome. Int J Radiat Oncol Biol Phys 1993;27:10511056. Gorodetsky R, Sheskin J, Weinreb A. Iron, copper, and zinc concentrations in normal skin and in various nonmalignant and malignant lesions. Int J Dermatol 1986(25):440444. Gutteridge JM. Iron and oxygen: a biologically damaging mixture. Acta Paediatr Scand Suppl 1989;361:7885. Hann HW, Stahlhut MW, Hann CL. Effect of iron and desferoxamine on cell growth and in vitro ferritin synthesis in human hepatoma cell lines. Hepatology 1990;11:566569. Heinz R, Lorant P, Stacher A. Alcohol and changes in the hematopoietic system; an epidemiological contribution to the problem of alcoholic anemia. Wien Kiln Wochenschr 1983;95:624628. Hockel M, Schlenger K, Mitze M, Schaffer U, Vaupel P. Hypoxia and radiation response in human tumours. Semin Radiat Oncol 1996;6:3 9. Ishiko O, Sugawa T, Tatsuta I, et al. Anemia-inducing substance (AIS) in advanced cancer: inhibitory effect of AIS on the function of erythrocytes and immunocompetent cells. Jpn J Cancer Res 1987;78:596606. Kanwar KC, Tikoo A. Hematological lesions in rat following heavy alcohol ingestion. J Environ Pathol Toxicol Oncol 1992;11:241245. Keer HN, Kozlowski JM, Tsai YC, Lee C, McEwan RN, Grayhack JT. Elevated transferrin receptor content in human prostate cancer cell lines assessed in vitro and in vivo. J Urol 1990;143:381385. Kopczynski Z, Kuzniak J, Thielemann A, Kaczmarek J, Rybczynska M. The biochemical modication of the erythrocyte membranes from women with ovarian cancer. Br J Cancer 1998;78:466471. Lin J. Immunohistochemical study of transferrin receptor expression in hepatocellular carcinoma. Chung Hua I Hsueh Tsa Chih (Taipei) 1992;72:8687. Ludwig H, Eritz E. Anemia in cancer patients. Semin Oncol 1998;25(Suppl 7):26. Mandishona E, MacPhail AP, Gordeuk VR, et al. Dietary iron overload as a risk factor for hepatocellular carcinoma in black Africans. Hepatology 1998;27:15631566. McCormack M, Nias AH, Smith E. Chronic anaemia, hyperbaric oxygen and tumour radiosensitivity. Br J Radiol 1990;63:752759. Miyamoto T. Study on cell proliferation of oral malignant tumor with transferrin receptor. Kokubyo Gakkai Zasshi 1992;59:2132. Moliterno AR, Spivak JL. Anemia of cancer. Hematol Oncol Clin North Am 1996;10:345363. Nordstrom B, Strang P. Microangiopathic hemolytic anemias (MAHA) in cancer. A case report and review. Anticancer Res 1993(13):18451849. Overgaard J, Hansen HS, Andersen AP, et al. Misonidazole combined with split-course radiotherapy in the treatment of invasive carcinoma of larynx and pharynx: report from the DAHANCA 2 study. Int J Radiat Oncol Biol Phys 1989;16:10651068. Pederson D, Sogaard H, Overgaard J, Bentzen SM. Prognostic value of pretreatment factors in patients with locally advanced carcinoma of the uterine cervix treated by radiotherapy alone. Acta Oncol 1995;34:787795. [38] Pilch J, Gierek T, Laskawiec J, Grzybek H. The use of X-ray microanalysis and transmission microscopy in studies of the chemical composition and ultrastructure of the microregions of precancerous tissues of the larynx. Otolaryngol Pol 1996;50:471478. [39] Quemener V, Bansard JY, Delamaire M, et al. Red blood cell polyamines, anaemia and tumour growth in the rat. Eur J Cancer 1996;32A:316321. [40] Rodger RS, Fletcher K, Fail BJ, Rahman H, Sviland L, Hamilton PJ. Factors inuencing haematological measurements in healthy adults. J Chronic Dis 1987(40):943947. [41] Shaper AG, Pocock SJ, Ashby D, Walker M, Whitehead TP. Biochemical and haematological response to alcohol intake. Ann Clin Biochem 1985;22:5061. [42] Shionoya K. A study of relationship between proliferative activity and expressed transferrin receptor content in cancer cells. Kokubyo Gakkai Zasshi 1994;61:580589. [43] Siegers CP, Bumann D, Trepkau HD, Schadwinkel B, Baretton D. Inuence of dietary iron overload on cell proliferation and intestinal tumorigenesis in mice. Cancer Lett 1992;65:245249. [44] Sokol RJ, Booker DJ, Stamps R. Erythrocyte antibodies, autoimmune haemolysis, and carcinoma. J Clin Pathol 1994;47:340343. [45] Soyer HP, Smolle J, Torne R, Kerl H. Transferrin receptor expression in normal skin and in various cutaneous tumors. J Cutan Pathol 1987;14:15. [46] Spivak JL. Cancer-related anemia: its causes and characteristics. Semin Oncol 1994;21(2 Suppl 3):38. [47] Stevens RG. Iron and the risk of cancer. Med Oncol Tumor Pharmacother 1990;7:177181. [48] Takeshi K, Katsuyuki K, Yoshiaki T, et al. Denitive radiotherapy combined with high-dose rate brachytherapy for stage III carcinoma of the uterine cervix: retrospective analysis of prognostic factors concerning patient characteristics and treatment parameters. Int J Radiat Biol Phys 1998;41:319327. [49] Tanaka N, Miyamoto T, Saito M, et al. Transferrin receptor expression in oral tumors. Bull Tokyo Med Dent Univ 1991;38:1926. [50] Teshima T, Chatani M, Hata K, Inoue T. High dose rate intracavitary therapy for carcinoma of the uterine cervix: I. General gures of survival and complication. Int J Radiat Oncol Biol Phys 1987;13:10351041. [51] Turobbaev MV, Osechinskii IV, Kubantseva IV, Eremenko LL, Chubarova MV. The dependence of hemoglobin and iron reserve indices on habitat conditions. Hematol Transfuziol 1989;34:4146. [52] van Asperen IA, Feskens EJ, Bowles CH, Kromhout D. Body iron stores and mortality due to cancer and ischemic heart disease: a 17year follow-up study of elderly men and women. Int J Epidemiol 1995;24:665670. [53] Whitney JF, Clark JM, Grifn TW, Gautam S, Leslie KO. Transferrin receptor expression in nonsmall cell lung cancer. Histopathologic and clinical correlates. Cancer 1995;1995:2025. [54] Wickramasinghe SN, Comdan B, Hasan R, Marjot OH. Correlation between acetaldehyde-modied haemoglobin, carbohydrate-decient transferrin (CDT) and haematological abnormalities in chronic alcoholism. Alcohol Alcohol 1994;29:415423. [55] Yoda J, Yamanaka N, Saito T, Samukawa T, Tamura S, Kawaguchi T. Characterization of cell lines from metastatic maxillary cancer. Nippon Jibiinkoka Gakkai Kaiho 1994;97(41):9429.

[19] [20] [21] [22] [23] [24] [25]

[26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36]

[37]