Epidemiologic and Clinical Characteristics of Thalassemia (Thal) Intermedia (TI) in the United States

Introduction

TI is increasingly prevalent in the US due to changing immigration patterns. It is underdiagnosed, leading to inadequate or delayed management. This study reviewed the prevalence, epidemiology, and clinical characteristics of TI in patients (pts) in the US.

Methods

Medical records from 1/1997- 4/2014 at 4 US hematology centers were reviewed. Index date was the 1st TI visit at a center on or after 1/1/1997. Eligible pts had a TI diagnosis (≤8 mean packed red blood cell transfusions per year (yr) over a ≥3-y period after index date) and ≥12 months of follow-up. Data spanned from index date to death or last record. Descriptive analyses of demographic and clinical data were done by TI subtype.

Results

Of 138 pts enrolled, 84 had α-thal, 39 had β-thal, and 15 had E/β-thal. 74% of α-thal pts had deletional (del) Hb H, and 26% had non-deletional (ndel) Hb H. 59% of β-thal pts had homozygous or compound heterozygous β-globin mutations (8% with α deletion, 51% without), 20% had a single β mutation with α-gene triplication, 13% had autosomal dominant β thal, and 8% had other β-globin mutations. Of the E/β-thal pts, 80% had E/β⁰ and 20% had E/β⁺.

Median age at index date was 2.3 yr (1.64 del; 6.1 ndel) in the α-thal group, 9.2 yr in the β-thal group, and 2.2 yr in the E/β-thal group. Most α-thal (77%) and E/β-thal (87%) pts were Asian; most β-thal pts were White (46%) or African-American (36%). Most α-thal (56%) and E/β-thal (53%) pts were of Southeast Asian origin; most β-thal pts were of Mediterranean (31%) or African (21%) origin. 21%, 10%, and 20% of α, β, and E/β-thal pts, respectively, were foreign-born, and 5%, 3% and 7%, respectively, were transfused outside of the US. Observation length was similar across subtypes (median: 5.3 yr).

Clinical comorbidities are shown in Table 1. 22% of pts received ≥1 transfusion, while 7% of pts received ≥8 transfusions in any 1 yr to treat anemia, acute hemolysis, or cardiac abnormality; increased transfusions were initiated due to growth failure, acute hemolysis, or unspecified reasons.

β-thal pts had a higher mean number of transfusions per pt per yr (PPPY) (α: 0.4 (0.0 del; 1.5 ndel), β: 0.9, E/β: 0.2) and higher mean serum ferritin (ng/mL) (204.3; 511.7; 362.4), and more often had iron chelation therapy (ICT) (11%; 28%; 7%). There was an association between higher serum ferritin, more frequent transfusions, and older age. In pts <10 yr, mean number of transfusions PPPY for regularly and ever transfused was 4.0 and 2.1, and serum ferritin for regularly, ever, and never transfused was 723.7, 438.0, and 146.4 ng/mL. In pts >18 yr, these values were 9.8 and 4.6 transfusions PPPY and 1166.8, 1042.2, and 521.4 ng/mL. An association also existed between ICT and higher mean serum ferritin (ICT: 769.9; no ICT: 463.7 ng/mL).

22% of pts had ≥1 liver iron test, and 18% had ≥1 cardiac iron test. Tested pts were older than those not tested (median yr, liver iron: 26.1 vs 7.4; cardiac: 20.1 vs 4.6). Among tested pts, 78% (25/32) had abnormal liver iron results. Median (range) LIC based on R2 or SQUID was 10.8 (2.5-18.2) mg/g dw, with corresponding within-12-month serum ferritin of 494 (127-1770) ng/mL. 49% (17/35) of pts tested had abnormal cardiac results based on electrocardiogram or echocardiogram.

Conclusion

TI in the US affects a diverse population. Our data showed a higher prevalence in African-Americans than previously documented. Rates of comorbidity and transfusion frequency increased with age. 18% and 4% of pts were born and transfused outside of the US, potentially leading to additional transfusion-related morbidity. Consistent with extant data, serum ferritin in TI often underestimated actual LIC, rendering more pts potentially eligible for ICT than observed. Morbidities observed in this study underscore the need for better and earlier diagnosis, substantiating the need for nationwide TI screening/surveillance to optimize management.