Patients with cold agglutinin disease (CAD) are more vulnerable to infectious agents, thus the COVID-19 pandemic has posed a particular risk to this population. Sutimlimab Phase 3 studies CARDINAL and CADENZA spanned the period before and during the pandemic; investigators were advised to vaccinate enrolled patients without stopping treatment. Of 61 completers from both studies, 47 received ≥1 dose of a COVID-19 vaccine. In the immunogenicity analysis (n = 27) all patients developed an immune response post-COVID-19 vaccination, with detectable immunoglobulin G anti-spike antibodies. Analysis of six patients with booster vaccinations demonstrated increased immune responses pre- to post-booster. COVID-19 vaccines were well tolerated in patients with CAD receiving sutimlimab treatment, and no signs of hemolytic exacerbations were observed post-vaccination.

Background:Immune-mediated thrombotic thrombocytopenic purpura (iTTP) may lead to microvascular thrombosis and mortality, despite patients receiving appropriate standard of care treatment (immunosuppressive therapy and therapeutic plasma exchange). Caplacizumab directly inhibits von Willebrand factor-platelet interaction and consequently prevents microthrombi formation. Objectives:This study aimed to determine the efficacy and safety of caplacizumab in diverse, clinically relevant patient subgroups. Methods:In this post hoc analysis of phase 3 HERCULES study (NCT02553317), patients were categorized by clinically relevant subgroups (prior iTTP history, iTTP severity at presentation, and initial immunosuppression regimen). Results:In patients with previous acute iTTP episodes, less severe disease at presentation, or those who received a corticosteroid-only initial immunosuppression regimen, time to platelet count response was shorter with caplacizumab vs placebo. Across all subgroups, fewer patients experienced a composite outcome of iTTP-related death, exacerbation, or major thromboembolic event on caplacizumab vs placebo. Placebo-treated patients remained at risk of exacerbations and refractoriness on either initial immunosuppression regimen (ie, corticosteroids only or corticosteroids plus rituximab). In the corticosteroids plus rituximab group, no exacerbations were reported in caplacizumab-treated patients, but 8 of the 16 (50%) patients experienced exacerbations in the placebo group. Safety outcomes were consistent with the findings of the main HERCULES study. Conclusion:Caplacizumab treatment of acute iTTP, in combination with therapeutic plasma exchange and immunosuppression, was safe and effective regardless of prior iTTP history, severity, or initial immunosuppression regimen and improved patient outcomes across clinically diverse subgroups. These findings emphasize the need for treatments with rapid onset of action that can reduce mortality and iTTP-related complications.

SummaryImmune thrombocytopenia (ITP) is an autoimmune disease typically associated with severely depleted platelet counts. However, additional symptoms (e.g. increased fatigue and memory/concentration difficulties) can profoundly impact patients' quality of life. The nature and severity of cognitive impairment in ITP, and potential association with patient/disease characteristics were evaluated in 49 adults with relapsed/refractory ITP. The Cogstate Brief Battery quantitatively assessed psychomotor function (DET), attention (IDN), visual learning (OCL) and working memory (ONB) individually, as well as DET/IDN and OCL/ONB composites. Clinically important cognitive impairment (defined as z‐score ≤ −1) for ≥2 individual tests was observed in 29 patients (59%). Impairment was highest for IDN (67% of patients), followed by DET (53%), ONB (39%) and OCL (16%). A higher magnitude of impairment was observed for the DET/IDN composite (mean z‐score −1.54; 95% CI, −1.94 to −1.13) than OCL/ONB (mean z‐score −0.21; 95% CI, −0.49 to 0.07). The severity of cognitive impairment was comparable to mild traumatic brain injury and associated with increasing age and fatigue but unrelated to platelet count or corticosteroid use. Overall, these results warrant a clinical need to further consider the potential of cognitive dysfunction in assessing ITP patients.

Topic: 28. Enzymopathies, membranopathies and other anemias Background: Cold agglutinin disease (CAD) is a rare, chronic autoimmune hemolytic anemia characterized by hemolysis from immunoglobulin M (IgM)-induced classical complement pathway activation. Sutimlimab, a first-in-class, humanized, monoclonal antibody selectively inhibits C1s of the C1 complex, thereby preventing classical complement pathway activation; sutimlimab is currently approved for use in the EU, Japan, and USA. The Phase 3 CARDINAL (NCT03347396) and CADENZA (NCT03347422) studies demonstrated that sutimlimab results in sustained improvements in anemia, hemolytic markers, and quality of life in patients with CAD; CARDINAL assessed patients with a recent history of transfusion (≥1 in the last 6 months), while CADENZA investigated patients with no recent history of transfusion (≤1 during the last 12 months, 0 during the last 6 months). Aims: This study assesses if baseline serum markers are associated with response as defined by the primary endpoints of the CARDINAL and CADENZA studies. Methods: All patients who received sutimlimab during Part A (26 weeks) of the CARDINAL (n=22) and CADENZA (n=19) trials were included in this combined analysis. Serum markers collected at baseline were evaluated for predicting responder status. Patients were classified as either responders or non-responders based on whether they met the primary endpoint criteria defined by the clinical trial. For a patient to be a responder in the CARDINAL trial, a hemoglobin (Hb) increase ≥2.0 g/dL from baseline at the treatment assessment timepoint (TAT; defined as the mean value from Weeks 23, 25, and 26) or Hb level ≥12.0 g/dL at TAT was required. For the CADENZA trial, a Hb increase ≥1.5 g/dL from baseline at TAT was required. Other mandatory conditions for both clinical trials for responder status were: no blood transfusion from Week 5 through Week 26, and no treatment for CAD beyond what was permitted per protocol. Descriptive statistics, frequency, or percentage were used to analyze outcomes. Results: Of the 41 patients included in the analysis, 29 (70.7%) were classified as responders (n=13 CARDINAL; n=16 CADENZA). The mean age (standard deviation [SD]) of responders (67.0 [9.9] years) was significantly lower than non-responders (74.3 [7.3] years; p-value=0.0262). No significant differences were noted for ethnicity, race, sex, geographic location, or body mass index. The median (range) reticulocyte count in responders (164.9 [28–301] 109/L) was significantly higher at baseline than in non-responders (102.0 [4–185] 109/L; p-value=0.0055). Significant differences in the median (range) reticulocyte index (semi-quantitative index of the adequacy of bone marrow response to anemia) between responders (4.7 [1–9]) and non-responders (2.6 [1–5]; p-value=0.0091) were also observed. Lower levels of median (range) IgM antibodies were noted in responders (2.37 [0.5–22.4 g/L]) compared with non-responders (6.22 [1.2–12.4] g/L; p-value=0.0080); upper limit of normal for IgM was 3.0 g/L. No significant differences in bilirubin, C4, CAD titer, erythropoietin, ferritin, Functional Assessment of Chronic Illness Therapy – Fatigue, haptoglobin, hemoglobin, lactate dehydrogenase, or total iron-binding capacity were noted at baseline between responders and non-responders. Summary/Conclusion: Post hoc analysis of Phase 3 trial data suggests that predictive serum biomarkers for sutimlimab response in patients with CAD may exist (such as reticulocyte count, reticulocyte index and IgM levels) and should be explored further. Figure:Keywords: Hemoglobin, Complement, Autoimmune hemolytic anemia (AIHA), Hemolysis

Topic: 32. Platelet disorders Background: The complex pathophysiology of immune thrombocytopenia (ITP) may contribute to an unpredictable clinical course and response to therapy. Treatment with oral Bruton tyrosine kinase inhibitor rilzabrutinib in adults with ITP in the global phase I/II trial (NCT03395210) showed rapid platelet response and continued durability with longer use, as well as an acceptable safety profile. Rilzabrutinib is known to impact multiple ITP pathways including reducing Fcγ receptor–mediated macrophage function and autoantibody production. Aims: Assess thrombopoietin (TPO) levels at baseline and after rilzabrutinib, and characterize the impact of TPO levels on platelet response following rilzabrutinib. Methods: Eligible ITP phase I/II study patients had 2 baseline platelet counts <30×109/L with required, but inadequate response to ≥1 prior/concomitant ITP therapy. Stable doses of concomitant TPO-receptor agonists (TPO-RA) and/or CS were allowed. Primary endpoints were safety and platelet response (≥2 consecutive platelet counts ≥50×109/L and increased ≥20×109/L from baseline without rescue medication). TPO/platelet levels were evaluated at baseline and weeks 12/24 using bivariate ANCOVA models adjusting for baseline TPO/platelet levels and TPO-RA use. Additionally, the effects of baseline TPO/platelet levels on the odds of being a responder/non-responder were obtained from univariate logistic regression models. Results: At baseline, 50 assessable ITP patients had a median age of 49 y (range, 19-74), 56% were female; n=18 received concomitant TPO-RA. Of 50 evaluable patients, 20 (40%) met the primary platelet response during the main treatment period. Among 32 patients with available data, baseline TPO levels were <100, 100-249, and ≥250 pg/mL in 17, 13, and 2 patients, respectively. Circulating median TPO levels were 94 pg/mL at baseline that decreased following rilzabrutinib to 78 pg/mL (week 12) and 69 pg/mL (week 24). Conversely, median platelet count at baseline of 14×109/L increased following rilzabrutinib at 12 and 24 weeks to 37×109/L and 41×109/L, respectively. Patients not receiving TPO-RA at baseline had lower platelet levels (–68.8×109/L) at week 24 vs those receiving baseline TPO-RA. ANCOVA models showed no relationship between either baseline platelet or TPO levels against platelet levels at either week 12/24 (Table). However, there was a positive relationship with higher baseline TPO levels associated with higher TPO levels at weeks 12/24. The logistic regression model showed that higher baseline platelet levels were associated with significantly greater odds of being a responder for every 1×109/L higher baseline platelet levels (OR=1.08 [95% CI, 1.00, 1.16]; P=0.04). Interestingly, elevated TPO levels were associated with reduced odds of being a responder although not statistically significant, suggesting that TPO levels may have no impact on response to rilzabrutinib. Responders and non-responders showed consistently normal levels of hemoglobin, neutrophils, reticulocytes, and reticulocytes/erythrocytes for a median of 100%, 94%-100%, 88%, 92%-93% of visits, respectively. There were no cases of neutropenia and 5 patients had anemia; none were treatment related and rilzabrutinib dosing was unchanged. Rilzabrutinib treatment was associated with stable B cell levels. All treatment-related adverse events were transient and grade 1/2. Summary/Conclusion: Following rilzabrutinib treatment, platelet levels increased over time, whereas serum TPO levels declined. ITP patients responded to rilzabrutinib irrespective of their baseline TPO levels.Keywords: Bruton’s tyrosine kinase inhibitor (BTKi), Platelet, Immune thrombocytopenia (ITP)

Introduction Efanesoctocog alfa (formerly BIVV001) is a new class of high sustained factor VIII (FVIII) replacement therapy, composed of a single recombinant FVIII protein fused to dimeric Fc protein, 2 XTEN polypeptides, and the D'D3 domain of von Willebrand factor (VWF), designed to overcome the endogenous VWF-imposed half-life ceiling. The XTEND-Kids (NCT04759131) study demonstrated once-weekly efanesoctocog alfa was well tolerated and provided effective treatment of and protection from bleeds in previously treated patients <12 years old with severe hemophilia A. Here, we provide a detailed report on efanesoctocog alfa as a treatment for bleeds during XTEND-Kids. Methods Previously treated patients <12 years old (≥150 exposure days [EDs] for patients 6 to <12 years of age and ≥50 EDs for patients <6 years old) with severe hemophilia A (<1 IU/dL [<1%] endogenous FVIII activity or a documented genotype associated with severe hemophilia) received efanesoctocog alfa 50 IU/kg once weekly for 52 weeks. In line with the study protocol, bleeds were treated with a single dose of efanesoctocog alfa (50 IU/kg). Additional doses (30 or 50 IU/kg, every 2-3 days) could be considered if needed. For minor/moderate bleeds occurring within 2-3 days of a recent prophylactic dose, a 30 IU/kg dose could also be used. The number and location of bleeds, as well as dose and number of efanesoctocog alfa injections required to resolve them were recorded. Patient assessment of the response to the first injection for treating a bleed was evaluated using the 4-point International Society on Thrombosis and Haemostasis (ISTH) scale (excellent, good, moderate, and none). Results Of 74 patients in XTEND-Kids, 73 were included in these post hoc analyses, including 38 subjects <6 years of age. One subject in the 6 to <12 group was excluded from these analyses as he did not receive weekly prophylaxis treatment as specified in the protocol, instead receiving intense consolidation treatment (2-3 injections per week for ~4 months) following 2 traumatic self-reported hip bleeds (however, no signs of bleeding or joint damage were observed by magnetic resonance imaging during the consolidation treatment period). The mean (SD) total number of EDs was 52.3 (6.83) for the overall population. The mean (95% confidence interval [CI]) estimated annualized bleed rate (ABR) was 0.61 (0.42-0.90). The mean (95% CI) annualized spontaneous, traumatic, and joint bleed rates were 0.16 (0.08-0.31), 0.40 (0.24-0.65), and 0.30 (0.16-0.57), respectively. Eighty-six percent of participants (n=63) had ≤1 bleed, including 64% (n=47) with no overall bleeds; 88% (n=64) had no spontaneous bleeds and 84% (n=61) had no joint bleeds. There were 43 bleeds treated with efanesoctocog alfa, of which 11 were reported as spontaneous (26%), 28 as traumatic (65%), and 4 (9%) were of unknown etiology (Table). The most common locations for bleeds were the joints (n=21, 49%) and skin/mucosa (n=15, 35%). The rate of bleeding events per week throughout the study remained consistently low (Figure). The mean (SD) duration between a treated bleed and nearest injection taken prior to that bleed was 4.7 (1.95) days, with a mean of 4.5 (1.56) days for a spontaneous bleed and 4.8 (2.12) for traumatic bleeds. A single injection was sufficient to resolve 95% of bleeds, and no bleed required more than 2 injections for resolution. The median (interquartile range) total dose required to resolve a bleed was 52.6 (50.0-55.8) IU/kg. Among first injections for bleeding episodes with an evaluation of response by participants (n=37), the vast majority (n=36, 97%) were rated as excellent or good. Conclusions A single dose of efanesoctocog alfa 50 IU/kg resolved 95% of bleeds in children <12 years of age, regardless of type or location. The hemostatic efficacy of efanesoctocog alfa was rated as excellent or good for nearly all evaluated first injections (97%). Once-weekly efanesoctocog alfa provided efficient prophylaxis with no bleeds in 64% of participants, and consistently low ABRs among all types and locations of bleeds. Efanesoctocog alfa provided effective treatment and prevention of bleeds in previously treated children <12 years of age with severe hemophilia A.

Background: Cold agglutinin disease (CAD) is a rare, chronic, autoimmune hemolytic anemia mediated by the classical complement pathway (CP).Sutimlimab is a first-in-class, humanized, monoclonal antibody that selectively inhibits C1s of the C1 complex, preventing CP activation and CP-mediated hemolysis. CARDINAL (NCT03347396) was an open-label, single-arm, Phase 3 study of sutimlimab in patients with CAD and recent history of transfusion; CADENZA (NCT03347422) was a randomized, double-blind, placebo-controlled, Phase 3 study of sutimlimab in patients with CAD and no recent history of transfusion. Both studies had a 26-week treatment period (Part A) followed by a long-term extension (Part B), reporting data up to 2 years after the last patient finished Part A of CARDINAL and up to 1 year after the last patient finished Part A of CADENZA. With long-term treatment in both studies, sutimlimab demonstrated sustained efficacy with improvements in hemolysis and anemia, sustained clinically meaningful improvements in quality of life, and it had a favorable safety profile. Aims: To report combined safety data from Part A and B of the Phase 3 CARDINAL and CADENZA studies in sutimlimab-treated patients with CAD. Methods: In CARDINAL Part A, patients received sutimlimab on Days 0 & 7, then biweekly until the end of Part B. In CADENZA Part A, patients received sutimlimab or placebo on Days 0 & 7, then biweekly; in Part B, patients continued to receive biweekly sutimlimab or switched from placebo to receive sutimlimab on Days 0 & 7, then biweekly. Data fromall enrolled patients who received at least 1 dose of sutimlimab in CARDINAL (N=24) and CADENZA (N=42), including a post-treatment follow-up 9 weeks after the last dose, were combined into the Safety Analysis Set. Endpoints for this analysis included the incidence of treatment-emergent adverse events (TEAEs), treatment-emergent serious adverse events (TESAEs), and adverse events of special interest (AESIs). AESIs were selected based on a list of important identified risks and important potential risks for sutimlimab. Results: The Safety Analysis Set included 66 patients. At baseline, the median age was 69.5 years (range 46-88) and the majority (72.7%) were female, with a median duration since CAD diagnosis of 5.7 years (range 0-33). The median patient follow-up was 129.1 weeks (range 5-175). Sixty-four (97.0%) patients experienced ≥1 TEAE ( Table 1). Eighty-six TEAEs assessed as related to sutimlimab by the investigator occurred in 30 (45.5%) patients. Fifty-three TESAEs were reported in 22 (33.3%) patients. Three (4.6%) patients each experienced one TESAE assessed as related or possibly related to sutimlimab by the investigator (related vitreous hemorrhage, related viral infection, and possibly related severe cerebral venous thrombosis). Seven patients discontinued sutimlimab treatment and/or the studies due to ≥1 TEAE, 4 in CARDINAL and 3 in CADENZA. Four patients died during the studies: 1 in CADENZA with a TESAE of lung squamous cell carcinoma and sutimlimab was withdrawn due to this TESAE prior to the patient's death; 3 in CARDINAL, 1 patient died after premature discontinuation of sutimlimab due to AEs (fatal infection of Klebsiella pneumoniae, acrocyanosis), 1 patient died due to a newly diagnosed hepatic cancer in the first month of the study, and 1 patient died due to exacerbation of CAD during the 9-week washout period approximately 1.5 months after the last dose of sutimlimab. No deaths were assessed as related to sutimlimab by the investigator. Treatment-emergent AESIs (selected based on a list of important identified risks and important potential risks for sutimlimab) are presented in Table 1. Fourteen TEAEs of serious infection were reported in 9 patients (13.6%; one assessed as related to sutimlimab by the investigator), 31 TEAEs of hypertension were reported in 17 patients (25.8%), 28 TEAEs of acrocyanosis and/or Raynaud's phenomenon were reported in 13 patients (19.7%), and 4 TEAEs of thromboembolic events were reported in 4 patients (6.1%). No TEAEs of serious hypersensitivity reaction and/or anaphylaxis, meningococcal infection or development of systemic lupus erythematosus were reported. Conclusion: The combined safety analysis of the Phase 3 CARDINAL and CADENZA studies (Part A and B) demonstrated that sutimlimab was generally well tolerated, with the type and frequency of TEAE consistent with an older and medically complex population.

Topic: 28. Enzymopathies, membranopathies and other anemias Background: Cold agglutinin disease (CAD) is a rare subtype of autoimmune hemolytic anemia, characterized by chronic hemolysis mediated by the classical complement pathway (CP). The ongoing COVID-19 pandemic has posed additional challenges in treating patients with CAD. Anti-CD20 therapies (used off-label in CAD) increase the risk of serious COVID-19 infection and reduce COVID-19 vaccine responses. Sutimlimab (SUT), approved by the FDA, PMDA and EMA for patients with CAD, is a first-in-class humanized monoclonal antibody which selectively targets C1s in the CP, addressing the underlying mechanism of hemolysis in CAD. The Phase 3 CARDINAL and CADENZA studies spanned the period before and during the pandemic, and investigators were advised to vaccinate enrolled patients without stopping SUT treatment. However, the impact of SUT on immune response to COVID‐19 vaccination remains unknown. Aims: To explore concomitant use of COVID-19 vaccines and SUT in patients with CAD. Methods: Post hoc analysis from the open-label extension (OLE) of the CADENZA and CARDINAL studies. Immunogenicity to COVID-19 vaccines was analyzed in a subset of patients fully vaccinated (2 doses) against COVID-19, and in those receiving a booster, who had consented to the use of stored samples (collected no later than 6 months after vaccination). Anti-SARS-CoV-2 Spike (EUROIMMUN Anti-SARS-CoV-2 ELISA) and anti-SARS-CoV-2 Nucleocapsid (Abbott Architect i2000 SR) IgG titers were measured. Data on tolerability to COVID-19 vaccine were collected through adverse event (AE) reports. Results: Of 61 completers from both studies, 47 received ≥1 dose of a COVID-19 vaccine; 14 received none; 11 had an additional booster. The mean inter-dose interval between first and second dose in fully vaccinated patients (n=37) was 38 days, and the mean time from second dose to booster was 165 days (range 85─208; n=11). The mean (SD) time from any vaccine dose to next SUT dose was 7.78 (4.03) days, and from last SUT dose to any vaccine dose, 8.85 (9.67) days (n=86 for both). Only 1/47 patients missed a SUT dose following the first vaccination, and 1 patient missed a SUT dose following booster administration. Both times were at the investigator’s discretion. No patients missed a SUT dose after the second COVID vaccination dose. Ten AEs were reported in 8 patients during the 7 days after any vaccine dose, but none more than once, and there were no serious AEs. No COVID-19 cases were reported during the OLE. The immunogenicity analysis comprised 27 patients; all developed an immune response post-vaccination, with detectable IgG anti-Spike antibodies. The mean time between last dose and serum collection was 65 days (range: 25–137). A further analysis of 6 patients with booster vaccinations demonstrated increased immune responses pre- to post-booster (mean [95% CI] log10 titer change: 1.82 [0.79–2.85]) (Figure), with significantly greater titer levels versus post second dose (p=0.0054). To rule out previous asymptomatic COVID-19 infection, anti-nucleocapsid SARS-Cov-2 IgG antibodies were assayed and were negative in all cases. Anemia and hemolysis markers were measured after vaccination; no indications of hemolytic exacerbation were observed. Summary/Conclusion: COVID-19 vaccination response was not impaired in patients receiving SUT for the treatment of CAD, and there was no need to modify the SUT dosing schedule. COVID-19 vaccines were well tolerated, and there were no signs of increased hemolytic marker activity following COVID-19 vaccination in patients receiving SUT.Keywords: COVID-19, Complement, Autoimmune hemolytic anemia (AIHA)

Objective: The management of exacerbations and disease relapse is important for patients with immune-mediated thrombotic thrombocytopenic purpura (iTTP). Severe ADAMTS13 (a disintegrin and metalloproteinase with a thrombospondin type 1 motif, member 13) deficiency during clinical remission is associated with risk of relapse and may guide prophylactic immune-modulatory therapy. We evaluated ADAMTS13 activity as a potential biomarker of exacerbation or relapse risk in the HERCULES and post-HERCULES studies. Methodology: This is a post hoc analysis of integrated data from the modified intent-to-treat (mITT) population of the Phase 3 HERCULES trial (NCT02553317) comparing caplacizumab and placebo (both plus standard-of-care treatment) in patients (pts) with iTTP and the 3-year follow-up post-HERCULES study (NCT02878603). ADAMTS13 activity was determined at baseline, weekly during treatment (post-TPE) and twice during follow-up. Recurrence risk was assessed according to ADAMTS13 activity, using TTP adverse event codes. Results: 49/144 (34%) pts in the HERCULES mITT had a recurrence during HERCULES or post-HERCULES. 140/144 pts had follow-up data after treatment end. Of these, 39 pts (28%) had a recurrence after treatment end; mean [SD] ADAMTS13 activity was 20.5% (28.7) in pts with recurrence vs 54.0% (34.9) in pts without; [P<0.0001]). ADAMTS13 activity was <20% at treatment end in 69.2% (27/39) and 27.1% (26/96) pts with/without recurrence (P<0.0001). Similar trends were seen across both treatment groups (Table). Conclusion: Regardless of the treatment received (caplacizumab or placebo), lower ADAMTS13 activity levels at end of treatment were associated with a higher risk of recurrence in the HERCULES and post-HERCULES studies. These data highlight the predictive value of ADAMTS13 levels on the risk of recurrence and may assist clinical decision-making in the treatment of iTTP. This content was first presented at ASH 2022 (abstract #2493).

Introduction: Immune thrombocytopenia (ITP) is a heterogeneous autoimmune disease characterized by autoantibody-mediated platelet destruction and impaired platelet production, resulting in thrombocytopenia and high bleeding risk. In patients with ITP, response to therapies has been challenging to predict, as neither durable responses nor long-term remissions are guaranteed. To date, there are several ITP therapies with varying efficacy and safety, but few predictive markers exist. Consequently, predicting the response to current therapies is crucial because it helps optimize the use of current and future therapies. Data from the ongoing, global phase 1/2 study (NCT03395210) provide evidence that rilzabrutinib, a Bruton tyrosine kinase inhibitor specifically developed to treat autoimmune disorders, could be a safe and efficacious treatment for patients with ITP. The aim of this report was to identify baseline factors that may predict response to treatment with rilzabrutinib in patients enrolled in the phase 1/2 ITP trial. Methods: A total of 11 baseline clinical variables and pharmacokinetics (PK) exposure at steady-state were selected as potential predictors of platelet response among 45 patients who initiated 400 mg rilzabrutinib twice daily (BID). Platelet response was defined as ≥2 consecutive platelet counts of ≥50x109/L, separated by ≥5 days, and an increase from baseline of ≥20x109/L without the use of rescue medication during the 4 weeks before the latest elevated platelet count, and early response was defined as having platelet counts ≥50×109/L by day 8. Clinical predictors were selected based on clinical considerations anticipated to be associated with platelet response and evaluated by univariate logistic regression models. For continuous variables, the t test was used to test for differences, whereas for categorical variables, the Fisher exact test was used. PK exposure at steady-state was estimated based on post hoc analysis from a previously established population PK model. Results: Of the 45 patients who initiated 400 mg BID rilzabrutinib, 18 achieved the primary platelet response, of whom 14 were early responders. Significant predictors of response to rilzabrutinib, based on the univariate logistic regression models, were shorter duration of ITP (P=.04) and no prior use of thrombopoietic receptor agonists (TPO-RA; P=.03) or rituximab (P=.02; Table 1). No prior rituximab therapy was also a significant predictor in early responders (P=.02; Table 1). Conversely, the use of concomitant ITP medication with TPO-RA or corticosteroids was not associated with predicting either response or early response to rilzabrutinib treatment. Similar results were noted upon assessment of continuous variables by mean difference. Compared with their counterparts, responders had a significantly shorter mean duration of ITP (P<.01) and lower number of prior lines of therapy (P=.01) as did the early responders (P<.01 for both; Figure 1). Across categorical variables assessed by comparing proportions of patients, significant associations with being a responder vs nonresponder were seen with no prior use of TPO-RA (P=.05) and lack of prior rituximab therapy (P=.03). Responders accounted for 29% vs 64% of patients with vs without prior TPO-RA and 23% vs 57% of those with vs without prior rituximab. No prior rituximab was also significantly associated with early response (P=.02) to rilzabrutinib, with 48% of these patients being early responders vs 14% of those who received prior rituximab. None of the PK exposure metrics significantly impacted response to rilzabrutinib. Conclusion: In heavily pretreated patients with ITP, the strongest predictors of response were short duration of ITP and no history of rituximab therapy, suggesting that patients who are more treatment naïve are more likely to respond to rilzabrutinib therapy than those receiving prior ITP medication. Nevertheless, interpretation of these observations should be caveated by their exploratory nature and the small sample size. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal