Simplified quantitative estimation in vitro of lymphocyte radiosensitivity applied to patients with chronic lymphocytic leukaemia.

Lymphocyte survival changes observed at 1, 2 and 3 days as responses to 3 doses of ionising radiation in vitro (40, 100, and 500 cGy) are analysed by computer according to a simple (single cell population) mathematical model. Intrinsic radiosensitivity, the susceptibility to lethal injury, which is expressed as the D37 value (the radiation dose permitting 37% survival), is estimated separately from the kinetics of subsequent death of lethally-irradiated cells (expressed as their half-life, or t1/2 value). Among the 35 patients with B-cell CLL studied (15 were never treated), both parameters varied widely and independently of one another. t1/2 ranged from 9-200 h and above, D37 from 14-500 cGy or above. Twenty-three patients were deemed 'radiosensitive' (D37 below 110 cGy). D37 level did not correlate with treatment status, mode of treatment, clinical staging (Rai) or lymphocyte count. With some exceptions, D37 remained relatively constant for individual patients with increasing duration of disease or alterations in treatment status. The assay method may prove useful as an aid in predicting response to low-dose splenic irradiation (SI) in CLL.

A preliminary study of low-dose splenic irradiation for the treatment of chronic lymphocytic and prolymphocytic leukaemias.

21 patients, 18 with chronic lymphocytic leukaemia (CLL) and 3 with prolymphocytic leukaemia (PLL), all B-lymphocyte origin and with progressive disease, were treated with a regime of low-dose splenic irradiation (SI). All patients experienced a rapid relief of disease-related symptoms. Following SI the total lymphocyte count (TLC) was markedly reduced in 18 patients. Partial to complete regression of splenomegaly occurred in 9 patients. Pre-existing anaemia of production failure type improved in 6 patients and as a result the haemoglobin rose to normal or near normal levels. SI caused the loss of T-, B- and Null-lymphocytes, but the loss of B-lymphocytes predominated. CLL/PLL became quiescent for long periods in 9 patients (CLL = 8; PLL = 1) but remained progressive in the other 9, all CLL. SI had no demonstrable effects on TLC, splenomegaly or anaemia in the 3 remaining patients (2 CLL, 1 PLL). The treatment was well tolerated by all, and side effects were almost absent. Transient reduction of neutrophils and platelets occurred commonly. No patient with initially normal neutrophil and platelet counts developed irreversible neutropenia or thrombocytopenia. In view of these effects, the ease of administration and the lack of side effects, further evaluation of low-dose SI, particularly in comparison with other regimes, seems worthwhile.

Primary thrombocythaemia treated with busulphan.

Thirty-seven patients with primary thrombocythaemia (PT) treated with busulphan have been followed for periods up to 25 years. Reduction of the platelet count to less than 400 X 10(9)/l resolved vascular occlusive symptoms, but haemorrhagic symptoms often remained unaltered. Cox regression analysis indicated that there were only two prognostically important presenting features; age had a strong inverse correlation with survival and vascular occlusive symptoms correlated with a better survival. Median duration of survival on treatment was 9 X 8 years. The number of deaths was 2 X 1 times that of a comparable control group, with deaths from myelofibrosis markedly increased. Deaths from thrombosis and malignant diseases, including leukaemia, were not significantly different from the number expected, which emphasizes the efficacy and the relative safety of busulphan for the long-term treatment of PT. Progression of PT into myelofibrosis occurred in 24% of cases and 9% became polycythaemic. Two additional cases of "thrombocythaemia' with a Philadelphia chromosome (and no overt evidence of chronic granulocytic leukaemia) are also presented.

The intrinsic radiosensitivity of lymphocytes in chronic lymphocytic leukaemia, quantitatively determined independently of cell death rate factors.

Survival curve shape for lymphocytes X-irradiated in vitro is governed by death rate as well as intrinsic radiosensitivity. We have resolved into these two components the survival curves obtained for CLL lymphocytes by use of a simple mathematical model. A multiple correlation coefficient comparing the predicted with the experimental survival curves was close to unity (0.954-0.999). For 14/18 patients with unequivocal B-cell CLL, the leukaemic (colchicine ultrasensitive) cells behaved as a homogeneous population (D37 0.32-1.28 Gy). This is similar to the more radiosensitive class of lymphocytes of normal blood (believed to include the B cells) and is some 4-fold less than the more radioresistant class (comprising most of the T cells). The lethally hit cells were homogeneous in death rate, which followed first order kinetics. The half-life (range 9-87 h) was, on average, some 50 per cent shorter than the more radiosensitive normal lymphocytes. The remaining four patients constituted a miscellaneous group. From one of these, it can be seen that an excessively slow death rate can give the misleading impression of radioresistance. It is hypothesized that the benefit afforded certain CLL patients treated with low-dose total body irradiation (TBI) or splenic irradiation (SI) may reside, partly, in the sparing of T lymphocytes of the helper type and in accompanying selective elimination (or functional inactivation) of those of the suppressor type.

Treatment of primary proliferative polycythaemia by venesection and low dose busulphan: retrospective study from one centre.

Sixty-five patients with primary proliferative polycythaemia (polycythaemia rubra vera) were followed during the period 1962-83 and analysed retrospectively. Primary control of PCV was by venesection only with low dose busulphan solely as required to keep the platelet count below 400 X 10(9)/l. Median survival was 11.1 years from diagnosis which is equal to or marginally better than with other reported regimens. Vascular causes of death were only a little higher than expected in a comparable normal population. Only deaths from acute leukaemia and myelofibrosis were significantly increased above the normal population incidence. There was no evidence to suggest that these transformations were busulphan induced. Analysis of the incidence of occlusive vascular lesions lends support to an earlier recommendation that the PCV level be maintained below 0.45. No support was found for the possible disadvantages of a predominantly venesection regimen, such as iron deficiency and reactive thrombocytosis. The case is put for this use of low dose busulphan. The data presented would warrant the future inclusion of this therapeutic regime as one limb of a controlled trial.

Hypersplenism and splenectomy in lymphoproliferative and myeloproliferative disorders.

Splenic pooling of red cells and an expanded plasma volume are considered to be among the major mechanisms responsible for the anaemia in hypersplenism. In those conditions in which massive splenomegaly is associated with various degrees of marrow failure, diagnosis of the cause of anaemia may be difficult. A simple technique was used to estimate the degree of hypersplenism, from red cell mass data, in 94 patients with unequivocal lymphoproliferative or myeloproliferative disorders. The splenic effect was found to correlate well with both the size of the spleen (r = 0.75-0.90) and the actual red cell mass (0.79), and was abolished by splenectomy. Clinical data is also presented on 43 of these patients who underwent splenectomy. The incidence and type of complications, survival figures, and possible criteria for patient selection are discussed.

Interpretation of measured red cell mass and plasma volume in males with elevated venous PCV values.

A method of interpretation of red cell mass (RCM) and plasma volume (PV) data is described. The results in 188 males with PCV's over 0.50 places the patients in 4 groups: true (absolute) polycythemias, relative (low plasma volume) polycythemias, high normal red cell mass (HNRCM) and 'physiological variant'. Absolute polycythemias were increasingly frequent at higher PCV levels but only reached 100% at a PCV of 0.60. They showed an 18% incidence in the lower PCV range of 0.500-0.519. Relative (low PV) polycythaemia was found in 18% of the patients with PCV values in the range 0.500 to 0.599. Although the HNRCM and 'physiological variant' types found mainly in the lower PCV ranges they occurred at the 0.54 level. While this method of interpretation of RCM and PV data is perhaps arbitrary, it does provide a basis for the proper study of the common group of patients with raised PCV levels in which classification, course and treatment are uncertain. In addition the findings suggest that both RCM and PV should be measured at all levels of PCV over 0.50; that relative (low PV) polycythaemia is a real entity but less common than sometimes believed; that diuretics do not have a notable part in its causation and that the common HNRCM, 'physiological variant' groups are incompletely understood and require further study.

Heterogeneity of T-lymphocyte chronic lymphatic leukaemia (CLL). Study with conventional surface markers and monoclonal antibodies.

Lymphocyte surface markers (E-SRBC, EAC, EA gamma and SmIg) and monoclonal antibodies (OKT3, OKT4, OKT8 and OKIa) were used to characterise the blood and bone marrow lymphocytes of T-cell CLL (8 patients). The diagnosis of T-cell CLL was made primarily as the majority of blood lymphocytes formed E-SRBC in each patient. Other markers-EAC, EA gamma and SmIg--showed different patterns of association with E-SRBC. These findings considered together described 4 different phenotypes amongst these patients: (a) E+ (3 patients), (b) E+, EAC+ (1 patient), (c) E+, EA gamma + (2 patients), and (d) E+, SmIg+ (2 patients). Similarly, 4 different groups were defined with the help of monoclonal antibodies. Helper T-cell (3 patients) and suppressor T-cell (1 patient) CLL showed predominantly helper T- and suppressor T-lymphocytes respectively. Mixed T-cell CLL (1 patient) comprised an equal proportion of both subpopulations, while the remaining 3 patients, with excess of one or other subpopulations along with a considerable proportion of Ia antigen-bearing lymphocytes, formed the indeterminate cell type CLL.

Measurement of K cell activity: use of human OR1R1 red cells and anti-D to study the role of different leucocytes as effector cells.

A method for quantitative assay of K cell activity (KCA) with an allogenic combination of human OR1R1 red cells and anti-D (IgG, incomplete antibody) as target cells is described. Lymphocytes alone have been shown to cause lysis of these target cells. Fc receptor-bearing lymphocytes appear to be the chief mediators of KCA. This sub-population is a heterogeneous mixture of sub-sets of other sub-populations defined on the basis of different surface markers and likely to show variable expression of Fc receptors by the EA gamma rosette method. Monocytes and neutrophils (and possibly eosinophils), most of which bear Fc receptors, are ineffective as effector cells. Monocytes may in some instances have an inhibitory effect on the expression of KCA of lymphocytes. The method described has the advantage of simplicity (easy availability of OR1R1 cells and anti-D) and a high degree of sensitivity (marked lysis at low L/T ratio) and appears suitable for routine use in clinical practice.

Colchicine ultrasensitivity of lymphocytes in chronic lymphocytic leukaemia.

It has been previously reported that, in comparison with normal lymphocytes, the lymphocytes in chronic lymphocytic leukaemia (CLL) are ultrasensitive in culture to the cytocidal action of colchicine. In this report the results of 240 colchicine studies in 87 patients with CLL are presented and analysed in terms of the diagnostic, clinical and haematological significance of colchicine ultrasensitivity (CUS) in CLL. All patients with CLL showed very significantly increased CUS of lymphocytes. The CUS did not vary with clinical and haematological status or time and only very rarely with treatment. There was no correlation between % CUS and subsequent course or survival. The results in CLL and in other forms of lymphocytosis and in non-Hodgkin's lymphoma are compared. It is concluded that the test is of considerable diagnostic value, particularly in identifying low-count and treated CLL and in the exclusion of non-CLL lymphocytosis. It appears to have no prognostic value but may have future application in monitoring treatment which selectively eliminates the abnormal cell population.

Erythroid colony formation in primary proliferative polycythaemia, idiopathic erythrocytosis and secondary polycythaemia: sensitivity to erythropoietic stimulating factors.

This is a study of in vitro production of erythropoietic colonies from peripheral blood and bone marrow of normal subjects and patients with different polycythaemic conditions. Proliferative stimuli included: 1. Fetal calf serum (FCS) only. 2. FCS plus a source of erythropoietic-stimulating activity (ESA). 3. FCS + ESA + erythropoietin (Ep). It was found that normal subjects and patients with secondary polycythaemia (SP) exhibited full colony growth only in the presence of both ESA and Ep, while patients with primary proliferative polycythaemia (PPP) showed colony production with FCS alone, further enhanced in the presence of ESA and Ep. A group of patients with idiopathic erythrocytosis (IE), namely with an increase of red cell mass not accompanied by other signs of myeloproliferative disorder, and without underlying cause, showed a heterogeneous response to ESA which in some patients was significantly greater than in normal subjects or in SP patients. It appears therefore that sensitivity of erythropoietic colony formation to Ep and ESA may be helpful in differentiating among various forms of polycythaemia; this study also establishes the heterogenicity of the IE group.

Viscosity and iron deficiency in treated polycythaemia.

An in vitro study of the effect of red-cell changes due to iron deficiency on whole-blood viscosity has been made on samples from patients with polycythaemia treated by venesection. When the PCV of the samples was adjusted to a standard value of 0.45, whole-blood viscosity was unrelated to decreasing MCH. In contrast, when samples were adjusted to a standard Hb concentration of 14 g/dl, whole-blood viscosity rose exponentially with decreasing MCH. This increase in whole-blood viscosity was shown to be a function of the increasing PCV that accompanies the falling MCH at this standard Hb value. It is suggested that in polycythaemia not due to hypoxia, hyperviscosity can be satisfactorily corrected by venesection controlled by monitoring the PCV alone. Other consideration may apply in hypoxic polycythaemia and these are discussed.

Cerebral blood flow and blood viscosity in patients with polycythaemia secondary to hypoxic lung disease.

Blood viscosity, cerebral blood flow (CBF) and cerebral oxygen carriage (CBF X arterial oxygen content) were measured in 12 patients with polycythaemia secondary to hypoxic lung disease. CBF and cerebral oxygen carriage were both significantly higher than in a comparative group of 20 patients with raised packed cell volumes and normal lung function. The patients with secondary polycythaemia then underwent venesection and their mean packed cell volume fell from 0.613 to 0.495. This led to a consistent reduction in blood viscosity, which fell by 44% at a low shear rate (0.67/s) and 33% at a high shear rate (0.91/s). CBF rose by 21% (p less than 0.01), but cerebral oxygen carriage did not significantly increase in the group as a whole. Four of the patients with secondary polycythaemia had complained of episodes of confusion before venesection, which improved considerably once the packed cell volume had been lowered. Headache was relieved in a further two patients and none of the subjects was adversely affected by venesection. It was not possible, however, to show a correlation between symptomatic improvement and an increase in cerebral oxygen carriage.

Effect of high haematocrit on alertness.

Patients with high-normal or above-normal haematocrit were found to have impaired alertness when compared with a control group matched for age and occupation. On retesting the controls had improved alertness scores attributable to a practice effect; but the patients, when retested after reduction of haematocrit by venesection, had improved significantly more than the controls. Improvement in alertness correlated very well with the increase in cerebral blood flow which followed venesection. Levels of venous haematocrit that are generally accepted as normal may not necessarily be optimum.

Viscosity, cerebral blood flow and haematocrit in patients with paraproteinaemia.

It has been suggested that blood viscosity is involved in the control of cerebral blood flow (CBF) (Thomas et al. 1977a, b, Humphrey et al. 1979). CBF, using the intravenous Xenon133 technique, blood viscosity and haematocrit were measured in 21 patients with elevated viscosity due to paraproteinaemia and found to be the same as in normal subjects. However, the paraproteinaemic patients were anaemic with a mean haematocrit of 0.342. This degree of anaemia would normally be associated with a high CBF. The paraproteinaemic patients were then compared with a group of 10 anaemic patients with matched haematocrits but without paraprotein bands. The whole blood and plasma viscosities were significantly higher in the paraproteinaemic patients and the CBF was significantly lower. The haematocrit, and therefore oxygen carriage, was similar in the two groups. It is likely that in the anaemia of paraproteinaemia the expected increase in CBF did not occur because of the limiting factor of increased viscosity. This suggests that in this instance viscosity rather than oxygen carriage is a major determinant in the control of CBF. This is further emphasized by the better correlation between CBF and blood viscosity than between CBF and haematocrit. It seems likely that viscosity and oxygen carriage are independent variables in the control of CBF.

Cerebral blood flow, haematocrit and viscosity in subjects with a high oxygen affinity haemoglobin variant.

The cerebral blood flow is low in primary polycythaemia, and it has been suggested that this is due to the increase in viscosity which accompanies the elevated haematocrit. In the present study cerebral blood flow has been measured by a non-invasive 133Xenon technique in six subjects with an elevated haematocrit secondary to a haemoglobin variant with increased oxygen affinity. Flow was significantly higher than normal and 81% higher than in 11 subjects of comparable age, matched for haematocrit and viscosity, but without the haemoglobin variant. In patients with this unusual type of polycythaemia, cerebral blood flow is high despite the elevated blood viscosity and the implications of these results are discussed.

Plasma and whole blood viscosity in treated primary polycythaemia.

Whole blood viscosity at a range of shear rates (230--0.77 s-1) and plasma viscosity have been measured in 39 patients with treated primary polycythaemia (idiopathic erythrocytosis and primary proliferative polycythaemia) and 30 age-matched normal individuals. There was a wider range of plasma viscosity and whole blood viscosity values at the same haematocrit (0.46), particularly at the lower shear rates, in the 'polycythaemic' group than the normal group. Hypochromic microcytic red cell changes present in 14 patients in the 'polycythaemic' group did not have any noticeable influence on whole blood viscosity at a given haematocrit value, since plasma protein factors override any possible effect of these red cell changes. The range of observed whole blood viscosity results in the 'polycythaemic' patients at the same haematocrit (0.46) was equivalent to the effect on whole blood viscosity of a rise in haematocrit from 0.41 to 0.51. Since there is such a range of whole blood viscosity at the same haematocrit, the haematocrit alone does not necessarily give a precise assessment of the viscosity of a whole blood sample. Since there is evidence from other publications that blood flow in vitro may be critically influenced by whole blood viscosity, this lack of precision should be considered when treating patients at risk of vascular occlusive episodes.

Cerebral blood-flow and viscosity in relative polycythaemia.

Cerebral blood flow (CBF), red-cell mass, and plasma volume were measured in 39 patients with venous haematocrit values in the range 0.47--0.58. Patients with true polycythaemia were thus excluded. The 39 patients studied fell into two groups--those with a measured red-cell mass in the high-normal range and those with normal red-cell-mass values but reduced plasma-volume values (the relative-polycythaemia, low-plasma-volume group). The mean CBF was low in both the high-normal red-cell-mass (HNRCM) group (45.8 ml/100 g/min) and low-plasma-volume (LPV) group (48.8 ml/100 g/min). The haematocrit in the HNRCM group was 0.504 and in the LPV group 0.513. In a control group of subjects with a mean haematocrit of 0.421 CBF was 68.6 ml/100 g/min. Venesection was associated with a significant rise in CBF in both groups of patients--to 59.7 ml/100 g/min in the HNRCM group and 65.0 ml/100 g/min in the LPV group. Whole-blood viscosity fell significantly in both groups. Both groups demonstrated significant inverse relationships between CBF and haematocrit and between CBF and blood viscosity.