Baseline or pretreatment quality of life (QoL) has been shown to be a prognostic indicator in different hematological malignancies. The data on prognostic significance of QoL in CML are limited. The aim of this study was to examine if baseline QoL is predictive for clinical response achievement in pts with chronic phase-chronic myeloid leukemia (CP-CML) who receive treatment with dasatinib as first or second line.
Total of 111 pts with CP-CML who participated in the multicenter observational prospective real-world study were involved in the analysis. All the pts were treated with dasatinib (100 mg daily) as first or second line in a real-world hematological setting. Clinical responses were assessed according to international clinical recommendations. For QoL assessment all the patients filled out the RAND SF-36 questionnaire. QoL impairment grade was evaluated for each patient. To examine baseline QoL as predictor of clinical response achievement (complete cytogenetic response, CCyR; major molecular response, MMR) univariate logistic regression was used. Backward Wald multivariate logistic regression to determine significant predictors of CCyR and MMR achievement during 12 mos of dasatinib treatment was performed with the following variables - age, gender, ECOG PS, time since diagnosis, line of TKI treatment, percent of Ph+ metaphases at start of dasatinib and baseline QoL impairment.
Out of 111 CP-CML pts 17 were TKI-naive, 23 - early switched to dasatinib after failure of imatinib treatment (median duration of imatinib treatment - 6 mos, range 1-8), 71 were imatinib-resistant or intolerant before second line treatment with dasatinib (median duration of imatinib treatment - 40 mos, range 4-121). The median age of pts was 48 y.o. (range 19-83), 52% of pts - males. Median of baseline percent of Ph+ metaphases was 95% (range 0-100, interquartile range 30.3-100); data were not available in 8% pts. At baseline 58% of patients exhibited no/mild QoL impairment (<25% QoL decrease from QoL population norm); 11% - moderate QoL impairment (25-50% QoL decrease from QoL population norm); 31% - severe/critical QoL impairment (>50% QoL decrease from QoL population norm). With a median follow-up of dasatinib treatment of 12 mos (range 1-13.8) the incidence of CCyR and MMR was 46%, median of achieving - 6 mos (2-12). Univariate logistic regression revealed that baseline QoL impairment grade is predictive for CCyR and MMR achievement (p<0.05). Multivariate logistic regression model showed that baseline QoL impairment grade, baseline percent of Ph+ metaphases and line of TKI treatment were significant predictors of CCyR and MMR achievement during dasatinib treatment. Neither age and gender, nor time since CP-CML diagnosis and ECOG PS were predictive for CCyR/MMR achievement (p>0.05). Our final model had a Nagelkerke R2 of 0.23 and correct classification rate of approximately 68% (p=0.001). The lower baseline percent of Ph+ metaphases, earlier start of dasatinib treatment and better baseline QoL predicted for the achievement of CCyR and MMR. Assuming similar percent of Ph+ metaphases and the same line of TKI treatment, pts with less baseline QoL impairment were 1.3 times (26%; 95% CI=13-43; p<0.05) more likely to achieve CCyR and MMR than pts with higher QoL impairment (QR=2.9, 95% CI=1.3-6.5; p<0.05).
In summary, the outcome of CP-CML pts treated with dasatinib could be predicted depending of baseline QoL data. The results of real-world data confirm that CP-CML pts with pronounced baseline QoL impairment have significantly worse clinical outcomes of dasatinib treatment in terms of optimal clinical response achievement. Thus, QoL data may not only be helpful in evaluating treatment outcomes from patients' perspectives but may also, like clinical information, be prognostic or predictive for clinical response, thus helping clinicians in their routine clinical practice to reach better decisions on patient management or identify their needs and decide on possible additional interventions, such as referral for counseling or psychosocial help and support.
Ionova: BMS: Research Funding.
Asterisk with author names denotes non-ASH members.