Cerebral Infarcts in Patients with Sickle Cell Disease Miguel R. Abboud, MD Professor of Pediatrics Hematology-Oncology Chairman, Department of Pediatrics and Adolescent Medicine American University of Beirut Medical Center Beirut, Lebanon

TermDefinitionImaging StrokeAcute neurologic syndrome that results from either vascular occlusion or haemorrhage, resulting in ischaemia and neurologic symptoms or signs lasting >24 hours Positive Transient ischaemic attack Acute neurologic syndrome with deficits lasting <24 hours Negative Silent infarctSmall infarct (typically <15 mm) evidenced by MRI but no neurologic deficits Area of increased signal on intermediate or T2-weighted MRI pulse sequences Definitions Adams RJ, et al. Hematology Am Soc Hematol Educ Program. 2001:31-46.

Stroke Subtype by Age Ischaemic stroke 1 –54% of cerebrovascular accidents –Highest in 1st decade and after 30 years –Peak incidence at 2–5 years Haemorrhagic stroke 1 –Highest in 2nd decade Silent stroke/infarct –Radiologic findings consistent with white matter disease 1 –10%–30% of patients with sickle cell disease (SCD) 1 –Associated with cognitive deficiencies 1 and higher stroke risk 2 1. Verduzco LA, et al. Blood. 2009;114: Miller ST, et al. J Pediatr. 2001;139:

●Multivariate predictors (P <.05 for each) 1 –Prior transient ischaemic attack (TIA): Relative risk (RR) = 56 –Anaemia: RR = 1.85 per 1 g/dL Hb decrease –Recent acute chest syndrome: RR = 7 –Acute chest syndrome rate: RR = 2.39 per event/year –Hypertension: RR = 1.31 per 10 mmHg increase ●Additional predictors –Silent infarcts: RR = 14 2 –Nocturnal hypoxia: Hazard ratio (HR) = 0.85 per 1% increase in O 2 saturation 3 1. Ohene-Frempong K, et al. Blood. 1998;91: Miller ST, et al. J Pediatr. 2001;139: Kirkham FJ, et al. Lancet. 2001;357: Risk Factors for Infarctive Stroke

Hulbert ML, et al. J Pediatr. 2006;149: /14 8/38 Stroke Recurrence Risk After Initial Simple vs Exchange Transfusion Transfusion Type All children received scheduled chronic blood transfusion therapy for at least 5 years after the first stroke and initial therapy RR = 5.0 (1.3–18.6; P =.02)

●Ischaemic stroke is treated with emergent simple or exchange blood transfusion 1 ●Without transfusion, 70% will recur within 2–3 years 1 ●With chronic transfusion, risk of recurrence is reduced by 90% 1 Management of Stroke and Prevention of Recurrence 1. Josephson CD, et al. Transfus Med Rev. 2007;21: Pegelow CH, et al. J Pediatr. 1995;126: Powars D, et al. Am J Med. 1978;65: Transfusion 2 No Transfusion 3 8/60 Study population with transfusion vs historical control subjects without transfusion 2 Cumulative observation time = patient-years 10/15

Aim of study: [hydroxyurea + phlebotomy = alternative arm] vs [transfusions + deferasirox = standard arm] for 30 months to prevent secondary stroke and reduce transfusional iron overload 161 paediatric patients with SCD and documented stroke and iron overload enrolled in SWiTCH 133 patients randomized 1:1 Alternative arm Hydroxyurea + phlebotomy n = 67 Standard arm Transfusions + deferasirox n = 66 Prediction: increased recurrence of stroke events in alternative arm but counterbalanced by better management of iron overload with phlebotomy Hydroxyurea for Secondary Stroke Prevention—SWiTCH Ware RE, et al. Blood. 2010;116:Abstract 844.

0/66 7/67 1. Ware RE, et al. Blood. 2010;116:Abstract NIH. Press release. June 4, Accessed 11/21/11 at: SWiTCH—Stroke Recurrence Higher with Hydroxyurea than with Transfusions 1 Study was terminated early 2 due to the marked increase in secondary stroke risk with hydroxyurea compared with transfusion therapy and no benefit of phlebotomy over chelation in reducing iron overload Transfusion + Deferasirox Hydroxyurea + Phlebotomy

Importance of Transcranial Doppler Screening in SCD Annual Stroke Risk ●Baseline risk from Cooperative Study of Sickle Cell Disease (CSSCD) is approximately 0.5%–1% 1 −If prior stroke, annual stroke risk is approximately 30% 2 ● Increased risk of infarctive stroke with TIA, lower baseline Hb, prior and recent acute chest syndrome (CSSCD study, no prior stroke), but yearly risk not quantitated 1 ● If abnormal transcranial Doppler (TCD), annual risk is 10%–13% per year 3 ● If MRI “silent lesions,” annual risk is approximately 2%–3% 4 ● Severe arterial lesions on angiography? −Assumed to be bad, 5 but yearly risk has not been quantitated 1. Ohene-Frempong K, et al. Blood. 1998;91: Powars D, et al. Am J Med. 1978;65: Adams RJ. Arch Neurol. 2007;64: Miller ST, et al. J Pediatr. 2001;139: Abboud MR, et al. Blood. 2011;118:

*Includes 1 patient with intracerebral haematoma. Adams RJ, et al. N Engl J Med. 1998;339: /130 11/67 P <.001 1/63 Median follow-up = 21.1 months Paediatric patients with SCD and abnormal TCD velocity were randomized to transfusion or standard care to prevent first stroke Stroke-Free Probability Is Increased with Long-Term Transfusions in Children with SCD

Early TCD Screening and Intervention ●Predictive factors and outcomes of cerebral vasculopathy in the Créteil newborn SCA cohort (n = 217, SS/Sβ 0 ) ●Screened with TCD early and yearly since 1992 ●MRI/MRA every 2 years after age 5 years (or earlier in case of abnormal TCD) ●Transfusions for abnormal TCD and/or stenoses ●Hydroxyurea to symptomatic patients with no macrovasculopathy ●Stem cell transplantation for those with HLA genoidentical donor ●Mean follow-up 7.7 years (1609 patient-years) Bernaudin F, et al. Blood. 2011;117:

Cumulative Risks in SCD Cohort with TCD Screening Cumulative risks by 18 years of age –Stroke: 1.9% (95% CI 0.6%–5.9%) compared with 11% –Abnormal: TCD 29.6% (95% CI 22.8%–38%) plateau at age 9 years –Stenosis: 22.6% (95% CI 15.0%–33.2%) –SI: 37.1% (95% CI 26.3%–50.7%) age 14 years All cerebral event risk by 14 years 49.9% (95% CI 40.5%–59.3%) Independent predictive factors for cerebral risk –Baseline reticulocytes count: HR per 1 x 10 9 /L increase –Lactate dehydrogenase: HR 2.78 per 1 IU/mL increase Conclusion: Early TCD screening and intensification therapy reduced risk of stroke by age 18 years from 11% to 1.9% –50% cumulative cerebral risk suggests more preventive intervention is needed Bernaudin F, et al. Blood. 2011;117:

TCD and Transfusions in Patients with Silent Infarcts—Conclusions Early TCD and transfusions effective in preventing strokes TCD does not screen for risk of silent infarcts Most patients who develop silent infarcts have normal TCD Different strategies needed Bernaudin F, et al. Blood. 2011;117:

●79 subjects having normalized TCD under transfusion were randomized 1 –38 to continue cRCT therapy –41 to discontinue cRCT therapy ●No neurologic events in the cRCT group 1 Abbreviations: cRCT, chronic red cell transfusion; TCD, transcranial Doppler. 1. Adams RJ, et al. N Engl J Med. 2005;353: NIH. Press release. December 5, Accessed 11/21/11 at: Graphic courtesy of Dr. Miguel R. Abboud. STOP II trial terminated after 2 years and concluded that it is unsafe to stop blood transfusions in patients who are at high risk of stroke 2 4.9% 34.1% Patients with Neurologic Events (%) STOP II Trial—Transfusion and Stroke Prevention

TransfusionStandard Care Total Stroke0 9*9 New or worse silent infarcts 066 No change P-value<.001 Pegelow CH, et al. Arch Neurol. 2001;58: STOP Trial—Transfusion Therapy vs Standard Care for Prevention of Secondary Silent Brain Infarcts *Includes 1 patient with new or worse lesion prior to stroke. Outcome after observation for 36 months in patients who had silent infarcts at baseline and who were randomized to transfusion or standard care

No Transfusion (n = 41) Baseline: Silent Infarct* 11 (26.8%) New Lesion 5 (50%) No Change 5 (50%) Baseline: Normal MRI 30 (73.2%) New Lesion 6 (20%) No Change 24 (80%) Transfusion (n = 38) Baseline: Silent Infarct 10 (26.3%) New Lesion 3 (30%) No Change † 7 (70%) Baseline: Normal MRI* 28 (73.7%) New Lesion 0 (0%) No Change 27 (100%) *1 patient had no follow-up MRI. † 3 patients had lesion number decrease; 1 reverting to normal scan. Abboud MR, et al. Blood. 2011;118: STOP II Trial—Effect of Discontinuing Transfusion on Silent Brain Infarcts on MRI

No Transfusion (n = 41) Baseline: Silent Infarct* 11 (26.8%) New Lesion 5 (50%) No Change 5 (50%) Baseline: Normal MRI 30 (73.2%) New Lesion 6 (20%) No Change 24 (80%) Transfusion (n = 38) Baseline: Silent Infarct 10 (26.3%) New Lesion 3 (30%) No Change † 7 (70%) Baseline: Normal MRI* 28 (73.7%) New Lesion 0 (0%) No Change 27 (100%) *1 patient had no follow-up MRI. † 3 patients had lesion number decrease; 1 reverting to normal scan. Abboud MR, et al. Blood. 2011;118: STOP II Trial—Effect of Discontinuing Transfusion on Silent Brain Infarcts on MRI At study end, 3/37 (8.1%) patients in the continued-transfusion group developed new brain MRI lesions compared with 11/40 (27.5%) in the transfusion-halted group (P =.03)

*Includes 1 patient with intracerebral haematoma. Adams RJ, et al. N Engl J Med. 1998;339: /130 11/67 P <.001 1/63 Median follow-up = 21.1 months Paediatric patients with SCD and abnormal TCD velocity were randomized to transfusion or standard care to prevent first stroke. Consequence of Stroke Prevention with Blood Transfusions IRON OVERLOAD is an inevitable consequence of chronic transfusions in patients with SCD Initial serum ferritin 164 ± 155 ng/L 1-year serum ferritin 1804 ± 773 ng/L 2-year serum ferritin 2509 ± 974 ng/L

Patient Characteristics GroupAge* (y) Transfusion Duration (y) Serum Ferritin (ng/mL) Liver Iron* (mg/g dry wt) SCD14.8 ± ± ± ± 1.38 β-Thal18.37 ± ± ± ± 2.15 Organ Dysfunction GroupCardiac Disease Growth Delay* Gonadal Failure SCD09%0 β-Thal20%27%33% Liver Disease GroupViral Hepatitis ALT >65 U/L Fibrosis Score >0 SCD2%7%39% β-Thal33%37%81% *P-value = not significant; P-value significant for all other comparisons. Vichinsky E, et al. Am J Hematol. 2005;80: Organ Dysfunction in Sickle Cell Disease and β-Thalassaemia SCD n = 43 β-Thal n = 30

●Organ injury may require a critical iron level with prolonged exposure ●SCD biology and its secondary inflammatory state may be protective factors ●Inflammation may decrease organ injury by restricting iron to shielded sites within the reticuloendothelial system (RES) and delaying the release of iron from the RES system ●The 2 diseases may have different transport and storage proteins Why Do SCD Patients Demonstrate Less Organ Injury than β-Thalassaemia Patients? Vichinsky E, et al. Am J Hematol. 2005;80:70-74.

Possible Explanations for Absence of Cardiac Iron Overload in SCD ●Nontransferrin-bound iron higher in thalassaemia major than SCD ●Other factors –Splenic tissue –Ineffective erythropoiesis –Gastrointestinal iron metabolism –Urinary iron loss Vichinsky E, et al. Am J Hematol. 2005;80:70-74.

Hepatocyte Siderosis Kupffer Cell Siderosis With permission from Pierre Brissot, MD.

How to Monitor Iron Status ●Serum ferritin –Noninvasive, available, inexpensive –Confounded by several parameters –Use long-term trends and avoid using acute-phase values ●Liver biopsy –Gold standard –Reveals pathology –Invasive –Sampling error ●Magnetic resonance –Accurate –Expensive

How to Manage Iron Overload ●Chelating agents –Desferrioxamine –Deferasirox –Deferiprone  Licensed for thalassaemia major only 1 ●Nonpharmacologic techniques –Erythrocytapheresis –Phlebotomy 1. Ferriprox (deferiprone). Summary of product characteristics. Leiden, Netherlands: Apotex; 1999.

Simple Transfusion 1 Manual Exchange Transfusion 1 Erythrocyta- pheresis 1,2 FeaturesEasy to perform 1 venous access Time-consuming Manual Expensive Requires 2 good venous accesses Good clinical tolerance Iron overload++++No iron overload SafetyAllo-immunization +++ Infections Chronic Transfusion Methods 1. Sickle Cell Society. Standards for the clinical care of adults with sickle cell disease in the UK Accessed 11/29/11 at: 2. Kim HC, et al. Blood. 1994;83:

PropertiesDesferrioxamine 1 Deferasirox 2 Deferiprone 3 Usual dose (mg/kg/d) 20–6020–3075 (total daily dose) RouteSC, IV (8–12 h, 5–7 d/wk) Oral, once dailyOral, TIW Half-life6 h8–16 h2–3 h ExcretionUrinary, faecalFaecalUrinary Key adverse effects Local reactions, ophthalmologic, auditory, growth retardation, allergic Gastrointestinal disturbances, rash, creatinine increase, ophthalmologic, auditory, elevated liver enzymes Gastrointestinal disturbances, agranulocytosis/ neutropaenia, arthralgia, elevated liver enzymes StatusLicensed for SCD Not licensed for SCD Iron Chelation Therapy is Needed to Treat Iron Overload 1. Desferal (desferrioxamine). Summary of product characteristics. Camberly, UK: Novartis; Exjade (deferasirox). Summary of product characteristics. Nuremberg, Germany: Novartis; Ferriprox (deferiprone). Summary of product characteristics. Leiden, Netherlands: Apotex; Graphic courtesy of Dr. Miguel R. Abboud.

Deferasirox (n = 117) Desferrioxamine (n = 56) Serum Ferritin Reduction P = NS Deferasirox vs Desferrioxamine— Measures of Iron Overload Data from Cochrane review of randomized-controlled trials that compared deferasirox with desferrioxamine. Abbreviations: LIC, liver iron concentration; SQUID, superconduction quantum interference device. Meerpohl JJ, et al. Cochrane Database Syst Rev. 2010;8:CD LIC Reduction (SQUID) P = NS Deferasirox (n = 83) Desferrioxamine (n = 33)

MeasureN Risk Ratio: Deferasirox vs Desferrioxamine Patient satisfaction (95% CI 1.99–4.93) Convenience (95% CI 2.28–6.47) Patient’s estimate of likelihood to continue treatment (95% CI 3.38–13.91) Discontinuations (95% CI 0.56–2.44) Data from Cochrane review of randomized-controlled trials that compared deferasirox with desferrioxamine. Meerpohl JJ, et al. Cochrane Database Syst Rev. 2010;8:CD Deferasirox vs Desferrioxamine— Measures of Satisfaction and Adherence

Conclusions Infarctive strokes are a devastating complication of SCD Chronic transfusion regimens are very effective in preventing stroke recurrence as well as new strokes in patients with abnormal transcranial Doppler Early transfusions seem effective in preventing development and progression of silent infarcts Iron accumulation in sickle cell disease is different compared with thalassaemia Iron chelators are effective in preventing iron overload in these patients