Treatment of Smoldering CLL

The French group proposed a classification that segregated stage A into A' and A'' (11,15,23). Criteria for A' were hemoglobin level higher than 120 g/L and lymphocyte count lower than 30,000/mm3; and for A'' they were hemoglobin under 120 g/L and/or lymphocyte count higher than 30,000/mm3. The survival of these two groups was clearly different, with a 5-yr survival of 82% in the A' group and 62% in the A'' group. Interestingly, the survival of the A' group was very close to that of a sex- and age-matched French population (9,10,16).

In a further step, the 609 patients included in the CLL-80 trial were classified according to Binet stage A, the further stages A' and A'', stage 0, as well as Montserrat's proposal (17) to define smoldering CLL. A' accounted for 80% of stage A, smoldering CLL according to Montserrat accounted for 58%, and stage 0 accounted for 48%. Five-year survival rates were, respectively, 87, 88, and 89% for these three groups of patients, and 5-yr freedom of disease progression rates were, respectively, 75, 80, and 84% (16). In addition, the long-term results of the CLL-80 trials showed that death related to CLL was observed in 27% of Rai stage 0 patients (including 40% of stage A patients) and 29% of A' patients (including 80% of stage A patients). Therapy was required for 43% of patients in Rai stage 0 and for 51% of A' patients (10). Although the choice between different proposals remains arbitrary, a definition of smoldering CLL appears to be feasible. Whatever the definition used, survival rates of patients are close to those of the normal population. However, about 10% of patients with smoldering CLL will progress to stage C within 5 yr; more than 25% of these patients will die from CLL, and 50% of these patients will need treatment. These results further emphasize the need for a better understanding both of the mechanisms involved in surveillance of the leukemic population and of the cause of this disease, which hopefully should allow the development of new therapeutic and effective approaches.

Neither the Rai nor the Binet staging system can predict which patients among the good prognosis group will shift into progressive disease, and the present modifications of these staging systems (aiming at early identification) also fail to predict evolution adequately (10). Either we are still lacking adequate prognostic markers, or evolution to a more aggressive disease may result from occurrence of a second (unpredictable) malignant hit.

To address this question, we have studied the long-term evolution of patients included in the abstention arms of the CLL-80 and -85 trials from the French Cooperative Group on CLL. Table 3 depicts progression to stages B and C and treatment requirement for stage A patients included in the abstention arm. In more than two-thirds of patients who progressed to more aggressive stages or treatment requirement such evolution occurred during the first 3 yr, which could indicate (at least for these patients) that evolution is a consequence of the presence of a previously unidentified prognostic factor. However, the one-third of patients for whom evolution occurs following the initial 3 yr, must be considered; for at least some patients, a second malignant hit could be involved. Table 4 depicts the different clinical and hematological parameters that could significantly predict subsequent treatment requirement. Although higher tumor mass, as expressed by increased lymphocytosis, splenomegaly, and lymph node involvement, could help in predicting treatment requirement, the initial presence of these parameters does not result in further treatment prescription for one-third of these patients.

Unfortunately, with the exception of clinicohematological staging systems, i.e., the Rai and Binet classifications, most of the prognostic factors so far described have been assessed only in retrospective studies and with heterogeneous cohorts of patients in terms of prognostic factors and prescribed treatments.

Serum levels of P2-microglobulin and lactate dehydrogenase (LDH) can help predict disease activity (18). However, although these factors can help to predict disease progression, as shown in Table 5, they do not provide better information in survival terms than the A' and A'' substaging. Soluble CD23 in serum results from cleavage of membrane CD23 and provides a highly quantitative and reproducible method to gain insight into CD23 expression. Very high levels of the soluble form are found in CLL and appear to be closely related to clinical stage, overall survival, and disease activity. Interestingly, in stage A patients, high soluble CD23 levels, whether observed at baseline or during the course of the disease, predict a high probability of short-term progression (19). Therefore, monitoring soluble CD23 levels in CLL could be a valuable tool in helping physicians to delineate patients who should be treated promptly. Again, soluble CD23 levels clearly predict disease progression but do not predict survival better than the A' and A'' substaging (20).

The level of serum thymidine kinase (TK) in CLL has been shown to increase according to disease stage and to predict shortened survival. Moreover, in patients with indolent disease (i.e., Binet stage A or Rai stage 0), increase in serum TK anticipates the manifestations of disease progression for several months and then allows us to forecast transition from smoldering disease to a more aggressive course (21). Studies comparing TK levels and survival of A' and A'' patients have not been reported. Recognition of these abnormalities at diagnosis is therefore of considerable potential interest, as they could better predict short-term progression in patients with early-

Table 3

Progression to Stages B and C or Switch to Treatment (Tt switch) for Stage A Patients Initially Randomized in the Abstention Arm of CLL-80 and CLL-85 Trials From the French Cooperative Group

CLL-80 (308 patients) CLL-85 (466 patients)

Table 3

Progression to Stages B and C or Switch to Treatment (Tt switch) for Stage A Patients Initially Randomized in the Abstention Arm of CLL-80 and CLL-85 Trials From the French Cooperative Group

CLL-80 (308 patients) CLL-85 (466 patients)

Time (mo)

A to C (%)

Tt switch (%)

A to C (%)

Tt switch (%)

<12

15/4.9

36/11.7

23/5.0

50/10.7

<24

28/9.1

64/20.8

53/11.9

114/24.4

<36

40/13.3

97/31.6

63/13.5

146/31.3

<48

50/16.5

112/36.5

69/14.8

165/35.6

<60

55/18.1

125/40.7

79/17.0

180/38.8

<72

57/18.7

135/43.9

82/17.6

187/40.3

<84

57/18.7

138/44.9

83/17.8

191/41.2

>84 <132

61/19.8

158/54.6

Total

61/19.8

158/54.6

83/17.8

CLL-80, Stage A, Abstention Arm: Variables Predicting Treatment Requirement

Table 4

CLL-80, Stage A, Abstention Arm: Variables Predicting Treatment Requirement

Variable

Total (no.) (n = 30S)

Treated (%) (n = 15S)

Untreated (%) (n = 15O)

p-value

Lymphocytosis

262

52

48

< 30 x 109/L

Lymphocytosis

46

61

39

0.001

> 30 x 109/L

Splenomegaly yes

40

65

35

0.0095

Involved areas

0

130

42

58

1

110

55

45

2

68

70

30

0.0001

Stage

0

127

43

57

1

129

58

42

2 + 3

52

65

35

0.0001

Stage A'

246

51

49

Stage A''

62

60

40

Comparison Between A', A'', Substaging and Lactate Dehydrogenase (LDH) and ß2-Microglobulin Levels

Factorsa

Table 5

Comparison Between A', A'', Substaging and Lactate Dehydrogenase (LDH) and ß2-Microglobulin Levels

Factorsa

Parameter

O

1

2

1 or 2

A'

No. of patients Survival by yr (%)b

151

57

10

67

176

3

98

87

47

81

96

5

95

71

31

66

91

7

88

53

0

45

84

a Cutoff values were < 1.25 N for LDH and < 3 mg for P2-microglobulin. 0 factor, normal values a Cutoff values were < 1.25 N for LDH and < 3 mg for P2-microglobulin. 0 factor, normal values for bath; 1 factor, increased values of 1; 2 factors, increased values of both.

Survival was according to CLL-related deaths.

stage of CLL, which would allow the physician to plan treatment more precisely. However, recent reports have emphasized that the presence in the leukemic B-cells of some cytogenetic abnormalities (22) or the mutational profile of IgV genes or the expression of the CD38 marker (23-26) are better predictors of rapid progression and survival. Patients with del(11q) or del(17p) progress rapidly, respond poorly to therapy, and have short survival. In contrast, those with normal cyto-genetics or isolated del(13q) have a good prognosis (22). However, to detect cytogenetic abnormalities, fluorescence in situ hybridization (FISH) techniques, which are expensive and not widely available, are required.

Since the vast majority of CLL B-cells coexpress membrane IgM and IgD and this phenotype is characteristic of normal naive B-cells, it has been proposed that the normal counterpart of leukemic cells may be mantle-zone B-cells assumed to correspond to naive B-cells expressing unmutated V genes. However, more than half of CLL cases harbor somatic mutations of VH genes (23), as if they had matured in a lymphoid follicle. These recent results suggest that there are two types of CLL: one that arises from relatively less differentiated (immunologically naive) B-cells with unmutated heavy chain genes, and has a poor prognosis, and another that evolves from more differentiated B-cells (memory B-cells) with somatically mutated heavy chain genes and has a good prognosis (23-26). However, recent data derived from gene expression profiling analysis failed to distinguish unmutated and mutated cases clearly, favoring the view that all cases of CLL have a common cell origin and/or a common mechanism of malignant transformation (27,28). Since the ability to sequence IgV genes is not available in most laboratories, a valid and easily performed surrogate assay is desirable. Since CD38 surface expression is predictive of a worse prognosis, Damle et al. (24) suggested that CD38 determination might be a useful alternative In this regard, there is some correlation between the expression of the CD38 antigen on neoplastic lymphocytes and IgVH gene mutations, but this is far from absolute. Interestingly, CD38 expression appears to have prognostic value by itself.

It is unclear whether patients with Binet stage A or Rai indolent forms whose leukemic B-cells express unmutated V genes or deleterious chromosomal abnormalities like del(11q) deletion or del(17p) would benefit from early treatment. This possibility should be tested in a prospective clinical trial. By contrast, there is consensus that most patients displaying advanced CLL forms (i.e., Binet stage B or C, or Rai stage III or IV or progressive Rai stage I or II) whose life expectancy does not exceed 7 yr should be considered for early treatment (9).

Was this article helpful?

0 0

Responses

  • sarah
    What is definition of smouldering cll?
    8 years ago
  • senait
    What stages are smoldering in cll?
    8 years ago

Post a comment