Improved Survival for Patients with CLL in the Era of Combination Chemoimmunotherapy - a Danish Population Based Study

Background

The treatment of chronic lymphocytic leukemia (CLL) is in rapid transition. As single agents, Fludarabine (F) was shown to be superior to chlorambucil (Rai, NEJM, 2000). Soon after F with Cyclophosphamide (C), showed superiority to F alone (Eichhorst, Blood, 2006; Flinn, JCO, 2007, Catovsky, Lancet, 2007). Subsequently, the addition of CD20 antibody rituximab (R) to FC for the first time showed a survival benefit for fit patients in a clinical trial (Hallek, Lancet, 2010). Likewise, in unfit patients the addition of CD20 antibodies to chlorambucil led to increased overall survival (Goede, NEJM, 2014). Eventually, BCR-targeted treatment for patients harboring TP53 aberrations demonstrated promising results (Farooqui, Lancet Onc, 2015). Here we assess the impact of these successive changes of therapy on the survival of patients with CLL in a Danish population-based cohort.

Methods

We studied the survival of a population-based cohort of CLL patients reported to the Danish Cancer Registry 1978-2013. Patients were categorized according to year of diagnosis from 1978-1994, 1995-2000, 2001-2005 and 2006-2013. For each CLL patient, we randomly selected 50 controls from the general population matched on age, gender and municipality. Kaplan MeierÕs survival curves and Hazard ratios (HR) and 95% confidence interval (95%CI) for death since time of diagnosis /matching date for controls were calculated. Change in survival probability relative to the controls with stratification on gender, age and calendar period of diagnosis was assessed.

Results

In total, 10500 patients were diagnosed with CLL in Denmark from 1978 to 2013 as follows: 1978-1994: 4651, 1995-2000: 1622, 2001-2005: 1664 and 2006-2013: 2563. Thus, the reported incidence of CLL increased slightly after year 2000. Overall, we observed a continuously decreasing risk of death for patients with CLL compared to matched controls, with HRs from 3.47 (3.37 - 3.58) to 2.09 (1.96 - 2.24) for patients diagnosed 1978-1994 and 2005-2013, respectively. In inter group analyses, a significant stepwise reduction in risk of death was observed for each period (Figure 1). In all age groups and calendar periods, male patients had an inferior survival compared to female patients and younger patients survived longer than older patients (p<0.0001).

Discussion

A significant improvement in survival probability for patients with CLL over time was found. This coincides with the introduction of FCR as standard therapy for younger patients with CLL (approval by EMA in 2009, affecting the cohort diagnosed 2006-2013). Significant survival improvement was also observed in the 2001-2005 cohort, possibly due to a shift to combination chemotherapy. Also for elderly patients, otherwise expected to get less intensive treatment in part due to co-morbidities, the survival improved over time. This may be accounted for by the introduction of semi-intensive chemotherapy like bendamustine, reduced intensity FC(R) and more recently chlorambucil combined with CD20-targeting antibodies.

For the first time, we here present population-based data showing significant improvement in survival for patients with CLL in parallel with the introduction of new chemo-immunotherapeutic regimens into clinical practice. These data substantiate the reported increased survival for patients treated with chemo-immunotherapy in clinical studies. Further investigation and cross-reference with treatment databases are warranted in order to assess the impact of new targeted treatment options for CLL.

Figure 1.

View largeDownload slide

Overall survival for patients (70-79 and 50-59 years), lower four curves; upper four curves represent matched controls.

Figure 1.

View largeDownload slide

Overall survival for patients (70-79 and 50-59 years), lower four curves; upper four curves represent matched controls.

Close modal