الفهرس 
Chronic Myeloid LeukemiaOnly 14 pages are availabe for public view
 |  | from | 262 |  |  |
| | | from | 262 | | |
Abstract
By univariate analysis for 6 months results for each drug alone
1. 18 patients ( 22.8%) achieved on imatinib therapy and 4 patients (22.2%) on nilotinib achieved CCyR.
2. 6 patients (7.6%) on imatinib and only one patient (5.6%) on nilotinib achieved MMR.
3. 34 patients (43%) on imatinib and12 patients (66.7%) on nilotinib achieved EMR.
4. 18 patients (22.5%) on imatinib and 11 patients (55%) on nilotinib had an OR according to ELN guidelines.
5. 40 patients (50%) on imatinib and 8 patients (40%) on nilotinib are non responders according to the ELN guidelines.
Finally, for all risk scores there were no significant relation to 6 months (CCyR, MMR, EMR, OR and failure) for each drug alone by univariate analysis except 1-group A high risk patients by Sokal ; Sokal identifies high risk patients when predict 6 months- EMR .2-group B low risk patients by EUTOS; EUTOS can predict OR in the low risk patients.
Multivariate analysis for the effect of TKIs on the outcome
1. CHR: non significant relation
2. CCyR: non significant relation
3. EMR: non significant relation
4. OR: TKIs could acheive OR
• By multivariate analysis after adjustment of baseline Sokal , the effect of nilotinib is 4 times that of imatinib in achieving OR (Odds ratio 4.2) with fair predictive value evidenced by (AUC .706)
• By multivariate analysis after adjustment of baseline Hasford , the effect of nilotinib is 3.6 times that of imatinib in achieving OR (Odds ratio 3.63) with fair predictive value evidenced by (AUC .712)
• By multivariate analysis after adjustment of baseline EUTOS , the effect of nilotinib is 4.8 times that of imatinib in achieving OR (Odds ratio 4.878) with fair predictive value evidenced by (AUC .701)
5. Failure: failure is less likely to occur with TKIs therapy
A-Predictive power of Sokal
1. 6 months as an end point shows that Sokal identifies low and high risk patients when predict CCyR, EMR .
2. 6 months as an end point Sokal identifies failure in high risk category by multivariate analysis(Odds ratio 4.7) but AUC has fair predictive value evidenced by (AUC.796 ) (specificity 84.6%) under the ROC curve
3. High risk patients are less likely to achieve OR, AUC has fair predictive value evidenced by (AUC.706) (sensitivity 93.1%) under the ROC curve
So, Sokal could predict (CCyR &EMR) in low and high risk and failure in high risk patients.
B-Predictive power of Hasford
1- 6 months as an end point shows that Hasford identifies low and high risk patients when predict chr and EMR
2- Hasford predict failure in high risk patients by multivariate analysis (Odds ratio 11.6) but AUC has fair predictive value evidenced by (AUC.756 ) (specificity 96.2%) under the ROC curve
So, Hasford can predict (CHR & EMR) in low and high risk, and failure in high risk patients
C-Predictive power of EUTOS
High risk patients by EUTOS are less likely to achieve OR evidenced by (AUC.706 ) (specificity 87.3%) under the ROC curve and more liable for failure evidenced by (AUC.703 ) (specificity 86.5%) under the ROC curve.