Abstract

Introduction

Hydroxyurea (HU) is the only FDA-approved medication to treat adults with sickle cell anemia (hemoglobin SS disease, SCA). Two studies showed improved survival with long term HU therapy. Conversely, subsequent studies reported most patients continue to die by the fifth decade of life despite long term treatment. Further, no studies have shown that HU prevents organ dysfunction. In this work, we sought to assess the effect of HU dosing on survival and organ function in SCA in a cohort of patients at a single referral institution.

Methods

Adults with SCA enrolled on the Bethesda Sickle Cell Cohort Study between January 2001 and October 2012 were included in this retrospective analysis. The NIH Biomedical Translational Research Information System was used to electronically retrieve clinical notes, administration records, and medication orders. Computer-assisted review facilitated the extraction of HU dose histories. Survival, laboratory, and cardiopulmonary parameters were compared between enrollment and most recent follow-up based on HU status and fetal hemoglobin (HbF) response.

Results

Of 388 subjects with SCA, 254 (65%) were treated with HU at a median dose of 19.4 mg/kg/day. Mean follow-up was 3.58±3.53 years. Of subjects prescribed HU, 166 (65%) were on a therapeutic dose. Subjects taking HU had a significantly higher HbF (9.2 versus 6.8%, p=0.0003) and mean corpuscular volume (MCV, 96.2 versus 88.3fL, p<0.001) at their initial visit compared to those not taking HU. A similar pattern was seen at the last visit for HbF (12.0 versus 6.5%, p<0.0001) and MCV (99.2 versus 87.4fL, p<0.0001), suggesting that HU was administered for a prolonged period of time. Total hemoglobin was not different between groups at either visit (8.9 versus 9.0g/dL at initial visit, p=0.41 and 8.9 versus 8.8g/dL at most recent visit, p=0.66 in HU versus no HU patients, respectively).

There was no difference in overall survival based solely on HU treatment (p=0.11). However, subjects prescribed therapeutic HU doses (15-35 mg/kg/day, N=166) were more likely to be alive than subjects who never took HU (N=131, p=0.04). Survival was similar when comparing sub-therapeutic HU (N=46) to no HU (p=0.52) groups, and survival was also similar between subjects prescribed sub-therapeutic HU versus therapeutic HU (p=0.46). Multivariate analysis confirmed HU dose is independently associated with prolonged survival (p=0.006).

At enrollment, creatinine was lower (1.2 versus 0.8mg/dL, p=0.0016), total bilirubin was greater (3.3 versus 2.8mg/dL, p=0.02), and tricuspid regurgitant velocity was higher (2.6 versus 2.5m/s, p=0.03) in patients on HU. There were no significant differences in ejection fraction, brain natriuretic peptide, transaminases, alkaline phosphatase, or direct bilirubin. Further, there were no differences for any markers of organ function at the most recent follow-up.

To determine if the best HU responses were associated with improved organ function, subjects were divided into groups based on whether their maximum HbF was within the lowest or the highest quartile. The HU dosages were 24.1 versus 4.8mg/kg/day in the high versus low HbF groups, respectively (p<0.0001). Maximum HbF was 26.1% as compared to 1.5% in the low HbF group. Ejection fraction was higher and transaminases and creatinine were lower at both visits in subjects with the highest HbF levels (see Table). After adjusting for age, only creatinine remained lower in the high HbF group at enrollment (p=0.007).

 First Visit Last Visit 
Variable Low HbF High HbF P-value Low HbF High HbF P-value 
Ejection Fraction (%) 56.2 59.9 0.004 57.5 60.2 0.04 
Alanine Aminotransferase (U/L) 34.6 25.7 0.01 38.9 28.5 0.03 
Aspartate Aminotransferase (U/L) 52.0 41.4 0.04 54.7 33.7 <0.0001 
Creatinine 1.7 0.7 0.002 2.0 0.9 0.004 
 First Visit Last Visit 
Variable Low HbF High HbF P-value Low HbF High HbF P-value 
Ejection Fraction (%) 56.2 59.9 0.004 57.5 60.2 0.04 
Alanine Aminotransferase (U/L) 34.6 25.7 0.01 38.9 28.5 0.03 
Aspartate Aminotransferase (U/L) 52.0 41.4 0.04 54.7 33.7 <0.0001 
Creatinine 1.7 0.7 0.002 2.0 0.9 0.004 
Conclusions

HU therapy correlates with prolonged survival and preservation of organ function, but only when given in therapeutic dosages. These HU associations are not observed for sub-therapeutic doses less than 15 mg/kg/day. Patients should be treated with a maximum tolerated HU dose to achieve the highest possible HbF response, ideally before organ damage occurs. Additional studies are indicated to further confirm these findings.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract