Page 134 - Haematologica March 2020

M. Hoeks et al.
Figure 7. Serum creatinine levels (mmol/L) in chelated and non-chelated patients per check-up.
Reasons for cessation of iron chelation therapy
Information on reasons of cessation of ICT was not routinely recorded in the study. However, information about the deferasirox-treated patients was available for seven patients: fatigue and diarrhoea (1 patient), physi- cian’s choice (1 patient), economic reasons (1 patient), renal failure (1 patient), no effect (1 patient), dyspepsia (1 patient), and lower limb cramps and dosage change (1 patient).
Renal function
Non-chelated patients had slightly higher median crea- tinine values compared to chelated patients at time of eli- gibility [non-chelated: median 86 mmol/L (p10-p90: 61- 135); chelated: median 79 mmol/L (p10-p90: 59-107)]. Forty-four chelated patients had higher serum creatinine levels at the first visit after discontinuing chelation com- pared with creatinine levels at time of eligibility (P=0.02 for all chelating agents and P=0.03 for deferasiox-treated patients). Renal function in non-chelated patients increased similarly over time (Figure 7).
Discussion
The results of this study indicate that ICT may improve OS in transfusion-dependent lower-risk MDS patients (LR-MDS). Our results are in line with several previously reported studies.7-10,12,25-28 Some of these studies attempted to correct for confounding factors, but still suffered from confounding by indication. This generally results in an overestimation of the beneficial effect of ICT on OS in LR-MDS patients. So far, one randomized controlled trial has been reported on this subject. The randomized, place- bo-controlled, TELESTO trial29 evaluated the event-free survival (EFS) (a composite outcome, including non-fatal events related to cardiac and liver function, and transfor-
mation to AML or death) and safety of deferasirox versus placebo in low- and intermediate-1-risk MDS patients. This study demonstrated an EFS risk reduction of 36.4% in the deferasirox arm (P=0.015). However, there was no difference in median OS in the deferasirox-treated arm (HR 0.83, 95%CI: 0.54-1.28, P=0.200) when compared with placebo, but more than 50% of the placebo-treated patients switched to ICT after study treatment discontin- uation (the placebo drug). The results of the TELESTO study are in line with our results. However, the included patients may not represent ‘real-life’ elderly MDS patients with multiple comorbidities, as reflected by the mean age of 61 years old of the patients included in TELESTO study compared to the mean age of 70 years in the EUMDS Registry study. Furthermore, low accrual rates and the crossover to ICT after cessation of the place- bo, affected the statistical power of the TELESTO study.
Meanwhile, well-designed prospective observational data, reflecting, ‘real-life’ data, contribute to the better understanding of the effect of ICT on OS in LR-MDS patients. Recently, a study from the Canadian MDS reg- istry demonstrated a superior OS for 83 chelated patients compared to non-chelated patients (5.2 vs. 2.1 years; P<0.001).30 The patients in this study were selected at the onset of transfusion dependency. Chelated patients became transfusion-dependent at a much longer interval from diagnosis than non-chelated patients (median 18 vs. 6 months) and OS was calculated from the time of becoming transfusion-dependent. Even after matching, some incomparability between the two groups remained in factors like concurrent treatment, presence of ringed sideroblasts, and ferritin levels. Therefore, confounding cannot be excluded in this study. Nevertheless, their con- clusions are in accordance with our findings, supporting the probable beneficial effect of ICT on OS in LR-MDS patients.
648
haematologica | 2020; 105(3)