Effects of Sulforaphane Obtained from Broccoli Sprout Homogenate in Patients with Sickle Cell Disease (SCD)

BACKGROUND

Sickle cell disease (SCD) is the most common hemoglobinopathy worldwide, characterized by chronic complications due to ongoing vaso-occlusion and hemolysis. Previous studies have shown that red cells from individuals with sickle cell disease (HbSS) have reduced NRF2 expression levels, which contribute to decreased oxidative stress capacity and increased hemolysis (Sangokoya, et al. 2010 Blood). Additionally, Macari and Lowry have shown that in vitro NRF2 activation of erythroid progenitors results in induction of anti-oxidant stress response genes, as well as increased percent fetal hemoglobin (HbF), which is known to prevent sickling (2011 Blood). Therefore, we hypothesize that NRF2 activation in SCD patients has potential therapeutic benefits by simultaneously inducing HbF and increasing the anti-oxidative stress capacity of red cells. We proposed to activate NRF2 by using sulforaphane (SFN), a well-known natural product enriched in broccoli sprouts. We conducted an open-label, dose-escalation clinical trial for SCD patients to investigate the safety and physiological effects of NRF2 activation by SFN through ingestion of a broccoli sprout homogenate (BSH).

METHODS

Male and female adult patients (> 18 years) with either HbSS or HbSߺ thalassemia were enrolled at the Duke Comprehensive Sickle Cell Center adult clinic. Exclusion criteria: RBC transfusion or a change in hydroxyurea dose in the last three months, ongoing pregnancy, diabetes, or renal insufficiency (BUN >21 mg/dL and/or creatinine >1.4 mg/dL). Inclusion criteria: Hematocrit (Hct) ≥ 20% and Hb > 6.0 g/dL. Recruited subjects were instructed to avoid additional SFN-containing foods before and during the study period. Subjects ingested a thawed preparation of BSH once daily for 21 days to allow for repopulation of red cells during therapy. Tolerability, toxicity, and physiological effects of NRF2 activation were determined at pre-treatment baseline (day 0), on the last day of ingestion (day 21), and after a wash-out period (day 49). Five patients were recruited for each dose (50g, 100g), with the smaller dose having elicited no safety concerns.

RESULTS

No safety concerns were noted among the subjects at either dose. In both cohorts, there were no significant differences in the adverse events, pain scores, complete blood counts, complete metabolic profile, reticulocyte count, and LDH levels when comparing days 0 and 21. In the 50g cohort, there was an overall but not statistically significant increase of average HbF from 14.5% to 14.9% (p=0.0786) in all five patients from Day 0 to Day 21; analyses are incomplete for the higher dose. We also observed a trend of NRF2 mRNA target gene induction, including heme oxygenase-1 (HO-1), NAD(P)H dehydrogenase (quinone 1, NQO1), and globin mRNAs, at day 21 vs day 0, which returned to baseline levels at day 49. In the 50g cohort, we observed a 66% increase of ho-1 and 44% increase of nqo1 mRNA levels at day 21 vs. baseline. In the 100g cohort, we observed a 14% increase of ho-1 and 42% increase of nqo1mRNA levels at day 21 vs. baseline.

CONCLUSION

Our pilot trial suggests that NRF2 activation by BSH may increase NRF2 expression programs and induce fetal hemoglobin. We aim to enroll more patients at escalating doses, as participants present a wide range of clinical variability and may show variable response. Additionally, the lack of statistical significance at the lowest doses along with a lack of safety concerns strongly compel us to pursue more potent NRF2 inducers to elicit more robust physiological changes for additional clinical trials.