Ortel, Thomas L.

The NCCN Guidelines for Cancer-Associated Venous Thromboembolic Disease provide strategies for the prevention, diagnosis, and treatment of venous thromboembolism (VTE) in adult patients with cancer. VTE is a common and life-threatening condition in patients with cancer, and its management often requires multidisciplinary efforts. The NCCN panel is comprised of specialists spanning various fields, including cardiology, hematology, medical oncology, internal medicine, interventional radiology, and pharmacology. The content featured in this issue specifically addresses the evaluation and recommended treatment options outlined in the NCCN Guidelines for the diverse subtypes of cancer-associated VTE.

Objective To develop new antiphospholipid syndrome (APS) classification criteria with high specificity for use in observational studies and trials, jointly supported by the American College of Rheumatology (ACR) and EULAR. Methods This international multidisciplinary initiative included 4 phases: 1) Phase I, criteria generation by surveys and literature review; 2) Phase II, criteria reduction by modified Delphi and nominal group technique exercises; 3) Phase III, criteria definition, further reduction with the guidance of real‐world patient scenarios, and weighting via consensus‐based multicriteria decision analysis, and threshold identification; and 4) Phase IV, validation using independent adjudicators’ consensus as the gold standard. Results The 2023 ACR/EULAR APS classification criteria include an entry criterion of at least one positive antiphospholipid antibody (aPL) test within 3 years of identification of an aPL‐associated clinical criterion, followed by additive weighted criteria (score range 1–7 points each) clustered into 6 clinical domains (macrovascular venous thromboembolism, macrovascular arterial thrombosis, microvascular, obstetric, cardiac valve, and hematologic) and 2 laboratory domains (lupus anticoagulant functional coagulation assays, and solid‐phase enzyme‐linked immunosorbent assays for IgG/IgM anticardiolipin and/or IgG/IgM anti–β 2 ‐glycoprotein I antibodies). Patients accumulating at least 3 points each from the clinical and laboratory domains are classified as having APS. In the validation cohort, the new APS criteria versus the 2006 revised Sapporo classification criteria had a specificity of 99% versu s 86%, and a sensitivity of 84% versus 99%. Conclusion These new ACR/EULAR APS classification criteria were developed using rigorous methodology with multidisciplinary international input. Hierarchically clustered, weighted, and risk‐stratified criteria reflect the current thinking about APS, providing high specificity and a strong foundation for future APS research.

Antiphospholipid syndrome (APS), an acquired autoimmune thrombophilia, is characterised by thrombosis and/or pregnancy morbidity in association with persistent antiphospholipid antibodies. The 16th International Congress on Antiphospholipid Antibodies Task Force on APS Treatment Trends reviewed the current status with regard to existing and novel treatment trends for APS, which is the focus of this Task Force report. The report addresses current treatments and developments since the last report, on the use of direct oral anticoagulants in patients with APS, antiplatelet agents, adjunctive therapies (hydroxychloroquine, statins and vitamin D), targeted treatment including rituximab, belimumab, and anti-TNF agents, complement inhibition and drugs based on peptides of beta-2-glycoprotein I. In addition, the report summarises potential new players, including coenzyme Q10, adenosine receptor agonists and adenosine potentiation. In each case, the report provides recommendations for clinicians, based on the current state of the art, and suggests a clinical research agenda. The initiation and development of appropriate clinical studies requires a focus on devising suitable outcome measures, including a disease activity index, an optimal damage index, and a specific quality of life index.

Abstract SERF-TTP is the first prospective case-control study to investigate epidemiology, risk factors & outcomes in patients with first episode of idiopathic TTP. METHODS: Each case of TTP is identified upon referral for therapeutic plasma exchange (TPE). Exclusion criteria include organ or allogeneic stem cell transplant, anti-neoplastic therapy or malignancy. 2 age & gender-matched controls are identified for each case. Epidemiologic information, lab data & plasma samples are collected from each case. Case & control data are collected by standardized interview. Relative risk & 95% CI computed by chi-squared tests. ADAMTS13 activity & inhibitor are centrally measured on pre-TPE samples. ADAMTS13 activity is measured by FRETS-vWF73 assay (Peptides Int.). ADAMTS13 inhibitor is assessed by Technozym ADAMTS13 INH ELISA (Technoclone). RESULTS: Data is available for 67 cases & 138 controls. The median age of cases is 40, & 82% are female. Medical history & exposures are shown in Table 1. Clinical & lab characteristics vary by ADAMTS13 activity (Table 2). Low titer ADAMTS13 inhibitors are detectable in some patients with normal ADAMTS13 activity. 30 day survival of cases is 95.7%. Adverse reactions to TPE occur in 59% of patients, most commonly allergic(85%) or citrate-related reactions(65%). 14.7% had a venous access complication, 50% were catheter-related thromboses. 10% of all adverse events required an ICU admission. CONCLUSIONS: Preliminary results from the case-control study suggest that predisposing factors for the development of TTP include recent infection or connective tissue disorder. Cardiovascular disease, prior history of venous thrombosis, clopidogrel use and lower income are more common in TTP cases than controls. Patients with normal ADAMTS13 activity are more likely to present with higher platelet count, abnormal renal function and neurologic symptoms. Case & Control Characteristics Cases (n=67) Controls (n=138) Relative Risk (95% CI) Medical History Connective Tissue Disease 11.9% 2.2% 5.4 (1.5–20.0) Cardiovascular Disease 13.4% 3.6% 3.7 (1.29–10.6) Prior Venous Thrombosis 13.4% 2.9% 4.6 (1.48–14.5) Infection (prior 2 wks) 32.8% 10.1% 3.2 (1.77–5.9) Medications (prior 12 wks) Antibiotics 31.3% 17.4% 1.8 (1.08–2.99) Clopidogrel 4.5% 0.7% 6.4 (0.65–62.9) Income Level < $25,000 42% 18% 2.3 (1.47–3.6) $25,000 – 70,000 37.3% 52.9% 0.7 (0.49–0.99) > $70,000 11.9% 24.6% 0.5 (0.24–0.98) Characteristics by ADAMTS13 activity ADAMTS13 activity N ADAMTS13 inhibitor (>15 u/ml) ADAMTS13 inhibitor titer (mean) Platelets (mean) Creatinine (mean) Neurologic symptoms Survival Comparison of ADAMTS13 ≤20% vs > 20%, p-values: * 0.0028, † 0.003, ‡ 0.73, # 0.48. < 5% 18 (37.5%) 100% 80.9 22,350 1.4 44% 94% 5–20% 4 (8.3%) 100% 76.4 20,000 1.1 75% 100% > 20% 26 (54.2%) 57.7% 22.5 58,810* 4.3† 58%‡ 89.5%#

Abstract Background : Many idiopathic thrombotic thrombocytopenia purpura (TTP) patients have severe deficiency of ADAMTS13, an enzyme that cleaves ultralarge von Willebrand multimers. We recently reported that thienopyridine-associated TTP is characterized by an immunologic pathway with severe ADAMTS13 deficiency and a non-immunologic pathway with higher ADAMTS13 activity levels. We now compare findings for idiopathic and thienopyridine-associated TTP patients. Methods : Clinical findings and laboratory findings were evaluated for 51 idiopathic and 39 thienopyridine-associated TTP. Results: Clinical findings were similar between idiopathic and thienopyridine-associated TTP for both severe ADAMTS13 deficient and non-deficient patients. Differences were noted in gender and age, relapse rates, and survival. Conclusion : Among TTP patients with ADAMTS13 deficiency, relapses are frequent in idiopathic TTP patients and Rituximab may be useful, while for thienopyridine-associated TTP patients spontaneous relapse are rare as long as no re-exposure occurs. Among ADAMTS13 non-deficient patients, survival is high following therapeutic plasma exchange (TPE) for idiopathic patients but not for thienopyridine-associated TTP patients. Despite similarities, idiopathic and thienopyridine associated TTP probably have different initiating factors. ADAMTS13 activity and clinical characteristics in idiopathic and thienopyridine-associated TTP Idiopathic severe ADAMTS13 deficiency (n=29) †Thienopyridine severe ADAMTS13 deficiency (n=26) Idiopathic non-severe ADAMTS13 deficiency (n=22) †Thienopyridine non-severe ADAMTS13 deficiency (n=13) *p<0.05 for comparison of severe ADAMTS13 deficiency versus non-severe ADAMTS13 deficiency. †Thienopyridine-associated TTP clinical and presenting neutralizing autoantibody data was summarized from the Journal of the American College of Cardiology publication by Bennett et al. in 2007. ‡Thienopyridine-associated TTP remission autoantibody data was summarized Tsai et al, 2002, Bennett et al, 2000, and Orimo et al, 2002, and included 3 patients. §Data from average of ADAMTS13 deficient and non-deficient cohorts. Mean Age 41 67 47 60 Sex (% female) 79% 54% 86% 39% Platelet count/mm3 (mean)* 18,978 9,269 51,409 35,000 Platelet count <20,000/mm3* 69% 96% 23% 38% Hemoglobin (mg/dl; mean) 8.5 8.3 9.0 8.4 Creatinine (mg/dl; mean)* 1.5 2.2 3.2 3.0 Creatinine >2.5 mg/dl* 11% 27% 48% 46% 30-day survival 96% 85% 90% 62% Long-term relapse* 42% 8% 0% 0% Neutralizing autoantibodies to ADAMTS13 (prior to TPE) § 79% 100% 27% 0% Neutralizing autoantibodies to ADAMTS 13 (measured at remission) § 55% 33%‡ 33% 0%

Abstract Background: Thrombotic thrombocytopenic purpura (TTP) is a rare disorder that presents with microangiopathic hemolytic anemia and thrombocytopenia, fevers, renal insufficiency and neurologic features. We reviewed clinical, laboratory, and outcome data for TTP cases with severely deficient versus non-severely deficient ADAMTS13 activity levels. Methods: Mean and median data were from the Surveillance, Epidemiology and Risk Factors for TTP (SERF-TTP) study group for idiopathic TTP cases, the Canadian Apheresis Group (CAG), and five published series (Zheng 2004, Raife 2004, Vesely 2003 (Oklahoma TTP-HUS Registry), Matsumoto 2004 (Japan Referral Center), Bennett 2007). Results: Compared to TTP cases with near-normal ADAMTS13 activity levels (n= 282), TTP cases with severe ADAMTS13-deficiency (n=185) were more likely to have severe thrombocytopenia, normal renal function and neutralizing ADAMTS13 antibodies. Severe ADAMTS13 deficient TTP cases have better overall survival after therapeutic plasma exchange (TPE) but are more likely to relapse. TTP patients with severe ADAMTS13 deficiency were primarily categorized as idiopathic or ticlopidine-associated, while TTP patients with non-severely deficient ADAMTS13 activity levels were frequently categorized as idiopathic, secondary to drugs (clopidogrel, quinine), stem cell transplantation, or cancer. Conclusions: Severe ADAMTS13 deficiency is most commonly idiopathic, has better survival following TPE, and a 35–40% spontaneous relapse rate. By contrast, non-ADAMTS13 deficient TTP cases are usually associated with an underlying disorder or external insults. Amongst this cohort, four series have 47–62% survival rates and three series, which contain mostly idiopathic cases, have 83–90% survival rates following TPE. From this, we propose that TTP may occur by three possible mechanism; ADAMTS13-deficient (antibody-mediated), an immunologic mediated pathway independent of ADAMTS13 (i.e. quinine) that is responsive to TPE, and endothelial injury related TTP that is unresponsive to TPE. Platelet count mean (x10^9/L) Creatinine mean (mg/dl) ADAMTS13 neutralizing antibodies (%) Survival % Relapse % * <15% ADAMTS13 activity cutoff Severe ADAMTS13 Deficiency (<10–15%) SERF-TTP (n=30) 19 1.3 83 97 41 Zheng (n=16) 19 1.6 44 81 38 Bennett (n=26) 15 85 Oklahoma (n=18) 12 1.8 94 81 38 Raife (n=50) * 13 1.2 92 35 Japan (n=34) 35 91 Canada (n=11) 16 2.4 82 Not Severely Deficient ADAMTS13 Activity (> 15%) SERF-TTP (n=22) 57 3.9 35 90 0 Raife (n=57) * 44 2.7 83 9 Canada (n=17) 57 4.1 88 Zheng (n=13) 40 3.0 0 54 Bennett (n=13) 62 Japan (n=66) 9 62 Oklahoma (n=94 ) 23 47 3

Abstract Thrombotic thrombocytopenic purpura (TTP) is a microvascular occlusive disorder characterized by systemic aggregation of platelets, thrombocytopenia, and mechanical injury to erythrocytes. Between 30% and 80% of TTP cases are associated with ADAMTS13 deficiency. Thienopyridine-derivative anti-platelet agents, ticlopidine and clopidogrel, are the drugs most commonly associated with TTP. The structures differ only by a carboxymethyl side-chain and have no common metabolites. Since 2002, our R01 research project has focused on evaluating thienopyridine-associated TTP. Herein, we present the final results. Clinical and laboratory data were obtained from case reports, the FDA’s MedWatch program, a Japanese national reference laboratory for ADAMTS13 assays, and apheresis centers at Duke University, University of North Carolina, Northwestern University, and the Mayo Clinic. Epidemiologic data for rate estimation for thienopyridine-associated TTP among persons who receive cardiac stents were obtained from international cardiology laboratories. Pharmacovigilance information was obtained from package inserts for the drugs. Most thienopyridine-associated TTP cases are associated with two weeks or more of ticlopidine rather than clopidogrel, are immune-mediated involving neutralizing antibodies to ADAMTS13, resolve with therapeutic plasma exchange (TPE), and have spontaneous relapses. Less frequently, cases are associated with clopidogrel, occur within days of drug initiation, may be a direct result of endothelial cell damage, are less responsive to TPE, and are less likely to recur. Thienopyridine-associated TTP patients with severe deficiency of ADAMTS13 activity have a different profile than those with normal ADAMTS13 levels. Among thienopyridine-associated TTP patients who have ADAMTS13 deficiency, TPE is usually performed for a few days and patients recover without detectable organ damage. In contrast, among thienopyridine-associated TTP patients who do not have ADAMTS13 deficiency, several weeks of TPE is required for recovery, and 30% mortality rates have been reported. Despite similar chemical structures, ticlopidine- and clopidogrel-associated TTP probably occur by different mechanisms and have different clinical presentations and expected outcomes. Clinical Characteristics Onset Platelet Count <20,000 (%) Creatinine ≤2 mg/dl (%) Neurologic Symptoms (%) Mortality With TPE/Without TPE (%) Relapse *p < 0.05 ticlopidine 10%* <2 weeks; 90% 2-12 weeks 84* 28* 28 15/44 Occasional clopidogrel 74%*≤2 weeks; 26% > 2 weeks 60* 55* 32 28/33 Rare Epidemiology, Pharmacovigilance & Basic Science Incidence (cases per treated patients) Safety ADAMTS13 Activity (%) ADAMTS13 Antibodies (%) ticlopidine 1:1,600 “Black Box” warning <10* 90* clopidogrel 1:100,000 package insert warning 50–100* 10*

Abstract Background: Compared to Whites, Black-Americans may have a 40% higher incidence of idiopathic VTE. However, whether other VTE characteristics vary by race is uncertain. Objective: To compare demographic and baseline characteristics among White- and Black-Americans with VTE. Methods: Using a standardized data-collection form, demographic and baseline characteristics were prospectively collected from consecutive consenting patients enrolled in seven Thrombosis and Hemostasis Centers from August 2003 to March 2008. For patients with objectively diagnosed VTE, demographic and baseline characteristics were compared among White- and Black-American VTE patients, both overall, and by age and gender. Results: Among 1960 White- and 368 Black-Americans with VTE, compared to Whites, Blacks had significantly less isolated DVT (73.9% vs. 86.5%, p<0.001) and significantly more PE ± DVT (45.1 vs. 39.6%, p=0.05). Blacks and Whites did not differ in mean age (43.4 vs. 44.3 years), but a significantly higher proportion of Black men VTE patients were in age groups 20–39 and 80+ years. Blacks and Whites differed significantly by gender, with a significantly higher proportion of Black women with VTE (69.8% vs. 59.1%, p=0.0001). Blacks had a significantly higher mean BMI (32.3 vs. 29.4 kg/m2, p<0.0001), a significantly lower proportion with recent surgery or trauma, active cancer and infection, and a significantly lower proportion with a family history of VTE or documented thrombophilia (solely due to reduced Factor V Leiden and Prothrombin G20210A prevalence). Conversely, Blacks had a significantly higher proportion with hypertension, hyperlipidemia, diabetes mellitus, chronic renal disease and dialysis, HIV and sickle cell disease. The two races did not differ significantly regarding the proportion with idiopathic VTE, CHF, CAD/MI, peripheral artery disease, stroke or other chronic neurological disease (i.e., cerebral palsy), chronic liver disease or autoimmune disease, including antiphospholipid syndrome. Compared to White women VTE patients, Black women VTE patients had a significantly lower proportion with recent surgery, trauma, infection, oral contraceptive use and possibly hormone therapy; and a significantly higher proportion with idiopathic VTE, hypertension, diabetes mellitus, chronic renal disease, HIV and sickle cell disease. Compared to White men VTE patients, Black men VTE patients did not differ significantly regarding recent surgery, trauma, cancer, infection or autoimmune disease, but Black men had a significantly higher proportion with hypertension, diabetes mellitus, HIV and sickle cell disease. Black and White men with VTE did not differ significantly regarding idiopathic VTE, CHF, CAD/MI, stroke, peripheral arterial disease and chronic renal disease. Conclusion: White- and Black-Americans with VTE differ significantly regarding demographic and baseline characteristics that may be risk factors for VTE.

Abstract Abstract 1159 Background. Heparin induced thrombocytopenia (HIT) is an immune disorder where platelets are activated by antibodies to a complex of platelet factor 4 antigen and heparin (PF4/H), leading to thrombocytopenia (HIT) and, potentially, thrombosis (HITT). Documentation of anti-PF4/H antibodies in addition to the appropriate clinical findings is essential for making a diagnosis of HIT. In the post-cardiac bypass surgery setting, however, the frequency of elevated anti-PF4/H antibodies is high, whereas the frequency of clinical HIT or HITT is relatively uncommon. Several studies have shown that the presence of anti-PF4/H antibodies may be associated with an increased frequency of adverse outcomes, even in the absence of clinical HIT. The primary objective of this study was to determine the relationship between a positive PF4/H antibody in the postoperative setting with adverse thromboembolic events occurring up to 3 months after cardiac surgery. Methods. Patients undergoing cardiac surgery who were not going to be treated with chronic anticoagulation postoperatively were eligible for this multi-center prospective cohort study. Data were collected daily during hospitalization, and then at 30 and 90 days after surgery using a structured interview format with a standardized questionnaire that included all thrombotic as well as hemorrhagic events, platelet counts, and utilization of antithrombotics in the postoperative setting. The primary outcome variable was a composite endpoint comprising arterial and venous thrombotic events and other miscellaneous events compatible with HIT, as well as death attributable to an event compatible with HIT. Citrated plasma was collected at baseline, pre-discharge (∼4–5 days after surgery), and the 30 day follow-up visit, processed, and stored at −80°C for testing. Laboratory analyses included an anti-PF4/H antibody ELISA (GTI, Waukesha, WI) on all samples, a high-heparin confirmatory test on samples with an OD reading >0.40, and a serotonin release assay (SRA) on all postoperative samples with an OD reading >0.40. A sample size of 800 patients was estimated in order to detect a 3% difference in thromboembolic events assuming a 2 to 10-fold increase risk attributable to seropositivity. Chi-squared testing was used to test the relationship between the primary outcome and postoperative anti-PF4/H levels. Results. Informed consent was obtained from 1030 eligible patients between August 2006 and May 2009, and laboratory and follow-up data were analyzable for 1016 patients. Thirty-day antibody data were available for 888 patients, and fully complete laboratory and 90-day follow-up data were available for 815 patients. The average age was 62 ± 12 years, and 73% of participants were male. A total of 769 patients underwent coronary artery bypass grafting and 237 underwent valve repair or replacement. During the entire study period, there were 17 (1.7%) deaths, 46 thromboembolic events in 44 patients (4.3%), and 25 hemorrhagic events in 24 patients (2.4%). Using an OD cutoff of 0.40 for the ELISA, 339 patients (33.4%) were positive for anti-PF4/H antibodies at the time of discharge, and 630 patients (62%) were positive by day 30. There was no correlation between seropositivity for anti-PF4/H antibodies at the day of discharge or at day 30 and the primary outcome (p=0.47 and 0.73, respectively). Incorporating the high-heparin confirmatory step did not improve the relationship between positive antibody results and the primary outcome. Using a higher cut-off value for the anti-PF4/H antibody ELISA of 1.0 decreased the number of patients with positive results (96 patients at the time of discharge [9.4%] and 221 patients at the 30-day follow-up visit [21.8%]), but this did not improve the relationship between antibody positivity at the day of discharge or day 30 and the primary clinical endpoint, since most patients with the primary endpoint had an ELISA OD below 1.0 (75th percentile of 0.90; 90th percentile of 1.22). Similarly, using the SRA did not identify a relationship between assay results and outcome. Conclusions. The presence of anti-PF4/H antibodies in the postoperative setting following cardiac bypass surgery is not associated with an increased risk for thromboembolic complications. Positive anti-PF4/H results in this clinical setting should be interpreted with caution and only in the context of clinical suspicion for HIT. Disclosures: Ortel: Instrumentation Laboratory: Consultancy; Eisai: Research Funding; GSK: Research Funding. Welsby:CSL Behring: Speaker; CSL Behring: Membership on an entity's Board of Directors or advisory committees; NovoNordisk: Principal Investigator. Heit:Daiichi Sankyo: Honoraria; Ortho-McNeil Janssen: Honoraria; Covidien: Honoraria.

Abstract Background: The Surveillance, Epidemiology & Risk Factors for TTP (SERF-TTP) study is the largest prospective cohort of idiopathic TTP cases to date. Methods: Patients with first episode of idiopathic TTP were enrolled at 11 sites in the US. Exclusion criteria include solid organ or allogeneic stem cell transplant, anti-neoplastic therapy or malignancy. All patients underwent therapeutic plasma exchange (TPE). Remission is defined as platelet count >150,000/mm3. ADAMTS13 activity was measured by fluorescence resonance energy transfer assay (FRETS-vWF73, Peptides Int.). ADAMTS13 inhibitor was assessed by ELISA (Technozym ADAMTS13 INH ELISA, Technoclone) and functional inhibition of normal ADAMTS13 activity (modified FRETS). Differences between the groups was evaluated by Chi-square test, t-test or Fisher’s exact test. Results: Complete data is available for 57 cases. 84% were female & median age was 42. ADAMTS13 was severely deficient (<10%) in 53%, moderate (10–50%) in 8%, & normal (>50%) in 39%. Adverse events were frequent & most commonly include citrate toxicity and allergic reactions. Long-term followup was available for 56 patients, and overall relapse rate was 25% with median time to relapse of 11 months. The relapse rate was 41% in patients with severe ADAMST13 deficiency & 0% in patients with normal ADAMTS13 activity (p=0.0077). Conclusions: Severe ADAMTS13 deficiency does not define all cases of idiopathic TTP, yet is associated with a unique syndrome characterized by severe thrombocytopenia, normal renal function, presence of ADAMTS13 neutralizing autoantibodies & high risk of relapse. Non-ADAMTS13 deficient idiopathic TTP has clinical features similar to thrombotic microangiopathies associated with stem cell transplant or drugs such as quinine, yet has a better than expected overall survival after TPE. There is likely an alternate disease mechanism for this cohort, possibly immunologic, which may be responsive to TPE and warrants further study. ADAMTS13 < 10% n=30 ADAMTS13 > 50% n=22 p-value * t-test, ** Fisher’s Exact Test, # Chi-Square Test Presenting Labs Hemoglobin (mean) 8.5 9.2 0.22* Platelet count (mean x10^9/L) 19 57 <0.0005* Serum creatinine (mean) 1.33 3.9 0.0012* ADAMTS13 antibody (neutralizing & non-neutralizing) % 100 63.6 0.0004** ADAMTS13 neutralizing antibody % 83 35 .0003# Presenting Symptoms Neurologic symptoms present 45 52.4 0.23# Diarrhea 23 30.4 0.79# Infection 40 43.5 0.79# Past Medical History Connective Tissue Disease % 10 8.5 0.28# Family history of TTP % 0 4.6 0.11# Remission Labs Hemoglobin (mean) 12.2 8.5 0.16* Platelet count (mean x10^9/L)) 169 152 0.15* Serum creatinine (mean) 1.02 3.7 0.0009* Outcomes Therapy-related Adverse events % 71 55 0.24# Time (days) to remission (plt > 150,000) 9.7 9.2 0.89* Early remission (< 8 days) % 73 57 0.20# 30 day Exacerbation rate % 35 28 0.63# 30 day Survival % 96.6 90.0 0.56** Long-term Relapse % 41 0 0.0077**