The prognostic value of ZAP-70, CD38 expression and IgVH somatic hypermutation(SHM) in CLL has been well documented. We investigated whether the proposed model of combining ZAP-70 and CD38 levels to identify patients likely to progress (Del Giudice et al 2005, Schroers et al 2005) remained valid when mutational status was considered and studied which combinations of these 3 parameters provided the most valuable prognostic information. ZAP-70 and CD38 were evaluated by flow cytometry and IgVH SHM was analysed by direct sequencing with 98% cut off. All 3 parameters were studied in 115 untreated CLL patients, 90% of which were advanced stage:(stage A stable 10%, stage A progressive 30%, stage B 37%, and stage C 23%). ZAP-70 and IgVH SHM showed 68% concordance, CD38 and IgVH SHM concordance of 69% and 75% of patients, using cut offs of ≥30% and ≥7%, respectively. The impact on time to first treatment /treatment free interval (TFI) for these parameters can be seen in Table 1.

Treatment Free Interval (TFI )and Prognostic Factors

No.of CasesMedian TFI (months)P Value
TFI=Time from diagnosis to date of first treatment 
Mutational status Umutated 68 23 0.00003 
 Mutated 47 61  
ZAP70 ≥20% Positive 37 24 0.00055 
 Negative 78 44  
CD38 ≥7% Positive 79 25 0.0005 
 Negative 36 61  
Mutation/ZAP70 ZAP70+/Unmutated 35 19 0.002 
 Discordants 36 25  
 ZAP70-/Mutated 44 64  
Mutation/CD38 ≥7% CD38+/unmutated 59 21 0.004 
 Discordants 29 37  
 CD38-/Mutated 27 77  
ZAP70/CD38 ≥7% ZAP70+/CD38+ 34 19 0.003 
 Discordants 48 39  
 ZAP70-/CD38- 33 72  
Mutation/ZAP70/CD38 ≥7% ZAP70+/CD38+/Unmutated 32 20 0.007 
 Discordants 57 30  
 ZAP70-/CD38-/Mutated 26 75  
No.of CasesMedian TFI (months)P Value
TFI=Time from diagnosis to date of first treatment 
Mutational status Umutated 68 23 0.00003 
 Mutated 47 61  
ZAP70 ≥20% Positive 37 24 0.00055 
 Negative 78 44  
CD38 ≥7% Positive 79 25 0.0005 
 Negative 36 61  
Mutation/ZAP70 ZAP70+/Unmutated 35 19 0.002 
 Discordants 36 25  
 ZAP70-/Mutated 44 64  
Mutation/CD38 ≥7% CD38+/unmutated 59 21 0.004 
 Discordants 29 37  
 CD38-/Mutated 27 77  
ZAP70/CD38 ≥7% ZAP70+/CD38+ 34 19 0.003 
 Discordants 48 39  
 ZAP70-/CD38- 33 72  
Mutation/ZAP70/CD38 ≥7% ZAP70+/CD38+/Unmutated 32 20 0.007 
 Discordants 57 30  
 ZAP70-/CD38-/Mutated 26 75  

Univariate analysis showed significance for TFI, for each variable. Regardless of the combinations used, 2 or all 3 variables provided significant prognostic information with respect to TFI. An intermediate prognostic group was identified for discordant cases. CD38 ≥7% proved a more significant value than ≥30% for this series, hence this cut off was used for subsequent analysis. IgVH SHM/CD38 provided the best discrimination between favourable and unfavourable prognostic groups, in relation to TFI, with the least number of discordants.

Concordant cases of CD38+/ZAP70+ were able to positively predict unmutated status in 94.1% of cases and ZAP70-/CD38- cases predicted mutated status in 78.8% of patients. The discordant ZAP70/CD38 cases could be further stratified by testing IgVH SHM (mutated cases median TFI: 42 m, unmutated cases median TFI: 19 m). Amongst the cases discordant for ZAP/CD38/IgVH SHM, the largest group (40%), was ZAP70 -/CD38+ /unmutated and showed median TFI:25m, comparable to the worst prognostic group for all combinations. There was no evidence of preferential IgVH gene usage in this discordant group. In conclusion we have shown that combining IgVH SHM/CD38 provides more refined prediction of TFI in this group. Combination of ZAP-70 and CD38 is useful for predicting time to first treatment without the need for IgVH SHM analysis, in concordant cases. Thus these simple tests( ZAP-70 and CD38), performed by flow cytometry continue to provide relevant prognostic information,although IgVH SHM is still the paradigm.

Author notes

Corresponding author

Sign in via your Institution