An allogeneic stem cell transplant (SCT) remains the only curative option for patients with β thalassemia major (TM). Conventional pre-transplant risk stratification is limited by its subjective nature of definition of hepatomegaly, adequacy of chelation therapy and also the need for an invasive liver biopsy. We have previously reported on the significant limitations in applying this risk stratification in our population (developing country) where the majority will fall into the high risk Class III group and further there is significant heterogeneity within this Class III subset that is not recognized by the conventional risk stratification (BBMT 2007;13:889). The major determinant of pre-transplant risk in patients with TM is related to iron overload, in organs such as the liver and heart. T2* magnetic resonance imaging (T2*MRI) has been widely utilized as a non-invasive measure of iron overload in these organs and has been reported to predict organ dysfunction and to guide adequacy of therapy. There is however no data available on the role of T2*MRI in pre-transplant risk assessment of patients with TM.

From Jan 2014 to June 2016, in 69 patients with TM undergoing an allo-SCT a pre-transplant a T2*MRI cardiac and hepatic value was generated using standard techniques and validated software. The median age of this cohort was 7 years (range: 2-20) and there were 43 (62.3%) males. Twenty four (34.8%) were Lucarelli Class I/II while 45 (65.2%) were Class III. Of the Class III patients 15 (33.3%) were Class III high risk (Class III-HR) as previously defined by us. All Class III patients received a treosulfan based reduced toxicity myelo-ablative conditioning regimen. In this cohort 5 (7.2%) had graft failure, 12 (17.4%) had acute GVHD, 10 (14.5%) had sinusoidal obstruction syndrome (SOS) and 5 (7.2%) patients died. The cause of death was related to GVHD in 4 and in one it was due to uncontrolled sepsis. The 3 year KM estimate of OS and TFS for this cohort was 91.4±4.0% and 84.7±4.9% respectively.

The median T2*MRI cardiac and hepatic value was 17.3 msec (range: 0.06-65) and 3 msec (range: 0.44-27) respectively. The correlation coefficient between T2*MRI cardiac and liver values was 0.169; P-value=0.164 (Fig 1A). It was of biological interest to note that there were patients in whom there was preferential hepatic or cardiac iron overloading as seen in Figure 1A. There was a trend to correlation with age and T2*MRI cardiac values (Fig 1C) similarly there was a significant though weak correlation between T2*MRI cardiac values and conventional risk stratification (Fig 1B). There was no significant correlation between either T2*MRI cardiac or hepatic values on graft rejection, SOS, GVHD or death nor did either of this parameters stand out as significant on linear regression analysis.

In conclusion, with currently used reduced toxicity regimens, in the absence of organ dysfunction there is limited prognostic values of T2*MRI cardiac and hepatic readings on immediate peri-transplant events. Its impact on long term follow up post-transplantation on organ function and its role in directing post-transplant chelation therapy remains to be analyzed. Further evaluation is also required to understand the genetic and environmental factors that favor iron over load in one over the other organ as clearly seen in a proportion of cases.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.