Dear Sir,
We previously found that the soluble transferrin receptor (sTfR) which fully reflects bone marrow erythropoietic activity, had an attractive diagnostic accuracy in predicting the risk of extramedullary haematopoiesis not only in patients with thalassemia intermedia1, recently redefined non-transfusion-dependent thalassaemia (NTDT), but also in those with thalassaemia major2. There are no studies linking sTfR levels to overall morbidity in patients with NTDT. In this study we examined the rate of disease-associated complications and relationships with sTfR levels in our series of patients subdivided into different genetic categories of NTDT. This was a cross-sectional study of all NTDT patients followed at the “Cardarelli” Hospital Microcythemia Centre in Naples, Italy. Among our population with NTDT, we selected only patients who had never received blood transfusions or had only occasionally been transfused because of surgery, pregnancy or concomitant illnesses.
All medical records of the patients with NTDT were analysed carefully to collect data on the presence of morbidities and complications common to NTDT; complications were defined according to criteria described in the OPTIMAL CARE study3. Since 2007, sTfR levels have been tested in our NTDT patients twice a year and far from any occasional transfusion administered. The need for iron chelation therapy was considered a complication. All these patients had been already evaluated for their genetic defects as previously described4 and α- and β-thalassaemia mutations were identified by a reverse hybridization assay (alpha and beta globin StripAssay, Nuclear Laser, Vienna, Austria). The 15 patients with haemoglobin H (HbH) were genotyped as having deletional HbH disease. sTfR levels were determined with a commercially available kit using the N Latex sTfR and BN II System (Siemens Healthcare Diagnostics, Marburg, Germany) nephelometric technique. The reference range is 0.76–1.76 mg/L. Each patient was also assigned a complication severity score simply by summing one point for each complication present. The patients were then categorised as having a low (complication score 0–1), intermediate (complication score 2), or high (complication score ≥3) complication severity score. To adjust for the worsening effect of age on morbidity5, one point was added for patients younger than 30 years and one point was subtracted for those older than 40 years. Overall, 96 patients were available for analysis. The study was approved by the Ethics Committee of “Cardarelli” Hospital, Naples. Statistical analyses were performed using MedCalc® (ver. 10.2.0.0, MedCalc software, bvba, Ostend, Belgium). Results for descriptive statistics were expressed as mean ± standard deviation. Fisher’s exact test was used to compare the incidence of different parameters between the four groups of patients. Student’s t-test was used to compare differences in parametric data. A p-value below 0.05 was considered statistically significant.
As shown in Table I, the haemoglobin levels were comparable among all groups. The nucleated red blood cell counts were comparable between patients with a homozygous or compound heterozygous state for β thalassaemia and those with a combination of a β defect plus a β chain variant, but they were significantly higher than those observed in the other two groups (p<0.01). In patients with HbH the levels of lactate dehydrogenase were significantly lower than those observed in the other groups (p<0.01). The group with triplicated α genotype associated with β heterozygosity received less occasional transfusions with respect to other groups but this difference was statistically significant only in comparison with the group with a homozygous or compound heterozygous state for β-thalassaemia; the percentage of patients with a homozygosity or compound heterozygosity for β-thalassaemia who were under chelation therapy was statistically significantly higher than that in the other groups (p<0.05). All groups showed comparable prevalences of patients with osteopenia and cholelithiasis; by contrast, extramedullary haematopoiesis, pulmonary hypertension, diabetes, hypothyroidism and hypogonadism, were more prevalent in both the group with a homozygous or compound heterozygous state for β-thalassaemia and the group with the combination of a β defect plus a β chain variant.
Table I.
Clinical and laboratory characteristics of the patients.
| Overall (n=96) | β/β compounds (n=45) | ααα/β (n=28) | β variant/β defect (n=8) | HbH (n=15) | |
|---|---|---|---|---|---|
| Median age, years (range) | 41.0 (17.9–78.8) | 43.4 (22–72) | 38.0 (18.1–78.8) | 41.3 (33.8–68.4) | 32.7 (17.9–63) |
| Male/female | 43/53 | 22/23 | 12/16 | 3/5 | 6/9 |
| 16–35 years, n (%) | 36 (37.5%) | 14 (31%) | 11 (39%) | 2 (25%) | 9 (60%) |
| >35 years, n (%) | 60 (62.5%) | 31 (69%) | 17 (61%) | 6 (75%) | 6 (40%) |
| Splenectomised, n (%) | 44 (46%) | 35 (78%) | 3(11%) | 4 (50%) | 2 (13%) |
| Haemoglobin, g/dL; mean±SD (range) | 9.6±1.0 (7.3–12.1) | 9.3±1.0 (7.3–11.5) | 9.8±1.2 (7.6–12.1) | 9.7±0.9 (8.3–11.0) | 9.9±0.9 (8.5–12.0) |
| Serum ferritin, μg/L | 478±420 (17–2,400) | 608±512 (124–2,400) | 409±269 (17–900) | 419±288 (76–912) | 237.7±212.6 (21–746) |
| Nucleated red blood cells ×109/L | 11,317±19,613 (0–100,000) | 20,342±22,145 (0–100,000) | 525±1,462 (0–7,200) | 19,153±28,385 (0–75,000) | 208.3±386.3 (0–1,300) |
| sTfR, mg/L | 7.7±3.9 (1.3–18.0) | 10.3±3.0 (5.7–18) | 4.7±2.1 (2.1–12.0) | 9.8±4.0 (2.8–17.0) | 4.6±1.6 (1.3–8.5) |
| Lactate dehydrogenase, U/L | 522±257 (218–1,881) | 582±205 (330–1,289) | 503±307 (272–1,881) | 608±380 (286–1,422) | 332.5±89.7 (218–569) |
| Creatinine, mg/dL | 0.70±0.20 (0.33–1.8) | 0.6±0.24 (0.33–1.8) | 0.7±0.16 (0.5–1.1) | 0.7±0.15 (0.4–0.81) | 0.7±0.09 (0.53–0.81) |
| Treatment | |||||
| Hydroxyurea, n (%) | 19 (20%) | 17 (38%) | 0 (0%) | 2 (25%) | 0 (0%) |
| Never transfused, n (%) | 37 (38.9%) | 12 (26.7%) | 15 (55.6%) | 2 (25%) | 8 (53.3%) |
| Occasionally transfused, n (%) | 58 (61.1%) | 33 (73.3%) | 12 (44.4%) | 6 (75%) | 7 (46.7%) |
| Iron chelation, n (%) | 48 (51%) | 35 (78%) | 9 (33%) | 3 (38%) | 1 (7%) |
| Complications | |||||
| Cholelithiasis, n (%) | 56 (60%) | 30 (67%) | 13 (50%) | 7 (88%) | 6 (40%) |
| Nephrolithiasis, n (%) | 17 (18%) | 11 (24%) | 2 (8%) | 3 (38%) | 1 (7%) |
| Osteoporosis, n (%) | 15 (17%) | 12 (29%) | 3 (12%) | 0 (0%) | 0 (0%) |
| Osteopenia, n (%) | 42 (47%) | 21 (51%) | 11 (42%) | 5 (63%) | 5 (33%) |
| Extramedullary haematopoiesis, n (%) | 28 (35%) | 23 (55%) | 1 (6%) | 3 (38%) | 1 (8%) |
| Hypothyroidism, n (%) | 15 (16%) | 10 (22%) | 4 (15%) | 1 (13%) | 0 (0%) |
| Thrombosis, n (%) | 10 (11%) | 8 (18%) | 2 (7%) | 0 (0%) | 0 (0%) |
| Hypogonadism, n (%) | 7 (8%) | 5 (12%) | 1 (4%) | 1 (13%) | 0 (0%) |
| Pulmonary hypertension, n (%) | 5 (5%) | 3 (7%) | 1 (4%) | 1 (13%) | 0 (0%) |
| Diabetes mellitus, n (%) | 2 (2%) | 2 (5%) | 0 (0%) | 0 (0%) | 0 (0%) |
SD: standard deviation; HbH: haemoglobin H; sTfR: soluble transferrin receptor.
Overall, the levels of sTfR varied from 1.3 to 18 mg/L (mean±SD: 7.7±3.9 mg/L) and were comparable among patients with a homozygous or compound heterozygous state for β-thalassaemia (mean±SD: 10.3±3.0 mg/L) and patients with the combination of a β defect plus a β chain variant (mean±SD: 9.8±4.0 mg/L) but they were statistically significantly higher than those observed in the other two groups (p<0.05). The area under the receiver operating characteristic curve was used to determine if any level of sTfR could be used to differentiate between patients with and without morbidity: above the cut-off of 7 mg/L, sTfR level showed an acceptable sensitivity and specificity in predicting almost all complications and particularly those more directly linked to erythropoiesis expansion such as osteoporosis, osteopenia, extramedullary haematopoiesis and thrombosis (data not shown).
As shown in Figure 1, there was a strong relationship (p<0.0001) between complication severity score and median change in sTfR levels.
Figure 1.
Box-bar plots comparing the correlation between sTfR levels and the previously defined complication severity score in the overall population.
Thirty-seven, 21 and 38 patients had a low (L), intermediate (I) and high (H) complication severity score, respectively. A Kruskal-Wallis test was conducted to evaluate differences in sTfR medians between groups: the test, which was corrected for tied ranks, was significant χ2 (2, n=96) = 29.045, p<0.0001.
sTfR: soluble transferrin receptor.
Our data indicate that the determination of sTfR level, a common and low-cost laboratory test, could contribute, irrespectively of the patient’s genotype, to correctly classifying the disease severity in the explored subtype of NTDT at the time of the study. However, this cross-sectional study suffers from several limitations beyond its intrinsic nature: firstly, some clinical pictures and differences emerge from the comparison of groups with restricted numbers of patients and/or differently aged patients; secondly, we could have missed or improperly considered same variables and/or complications in assessing the overall disease severity; thirdly, it is conceivable that the few patients who had been receiving treatment such as bisphosphonates and/or calcium + vitamin D supplementation for osteoporosis or foetal haemoglobin induction therapy (in the case of β-thalassaemia) were presumably protected from the development of several complications in the presence of unchanged serum marker levels. Unfortunately, we have only been collecting data on sTfR level for a few years for most patients; consequently, we are not able to compare these levels with those at diagnosis or to ascertain whether they may undergo longitudinal changes according to complication score or whether they should be checked regularly. Further prospective studies are needed to evaluate whether the evaluation of sTfR level at diagnosis could contribute to early identification of the course of disease directly linked to marrow erythropoietic activity and expansion and could, in practice, help to monitor disease progression. It could be hypothesised that the higher the sTfR level, the earlier the patient should start bone density monitoring and evaluation of the presence of both extramedullary haematopoiesis2 and significant iron overload. Further prospective studies are also needed to compare sTfR level with other recognised markers of disease severity such as LIC, GDF 15 and HBF and to evaluate whether sTfR level, alone and in combination with these and other parameters, could help to monitor disease progression and treatments.
Footnotes
The Authors declare no conflicts of interest.
References
- 1.Ricchi P, Ammirabile M, Costantini S, et al. A useful relationship between the presence of extramedullary erythropoeisis and the level of the soluble form of the transferrin receptor in a large cohort of adult patients with thalassemia intermedia: a prospective study. Ann Hematol. 2012;91:905–9. doi: 10.1007/s00277-011-1385-y. [DOI] [PubMed] [Google Scholar]
- 2.Ricchi P, Ammirabile M, Spasiano A, et al. Extramedullary haematopoiesis correlates with genotype and absence of cardiac iron overload in polytransfused adults with thalassaemia. Blood Transfus. 2014;12S:124S–30S. doi: 10.2450/2013.0287-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Taher AT, Musallam KM, Karimi M, et al. Overview on practices in thalassemia intermedia management aiming for lowering complication rates across a region of endemicity: the OPTIMAL CARE study. Blood. 2010;115:1886–92. doi: 10.1182/blood-2009-09-243154. [DOI] [PubMed] [Google Scholar]
- 4.Ricchi P, Ammirabile M, Spasiano A, et al. Hypocholesterolemia in adult patients with thalassemia: a link with the severity of genotype in thalassemia intermedia patients. Eur J Haematol. 2009;82:219–22. doi: 10.1111/j.1600-0609.2008.01195.x. [DOI] [PubMed] [Google Scholar]
- 5.Taher A, Musallam KM, El-Beshlawy A, et al. Age-related complications in treatment-naїve patients with thalassaemia intermedia. Br J Haematol. 2010;150:480–97. doi: 10.1111/j.1365-2141.2010.08220.x. [DOI] [PubMed] [Google Scholar]